Compare commits

..

1 Commits

Author SHA1 Message Date
will.anderson d9791f15da docs: add cognitive-architecture treatise (06)
The middle layer between the whitepaper (thesis) and the engram API
reference (surface): how the mind is designed and why, as tiered
subsystems (LIVE/STAGED/DESIGNED). Covers the geometry thesis, engram
durability and the two-store topology (with the cultivate write-through
gap flagged as a known issue), reified Neighborhood nodes, the body/orbit
model, the operator/language/interoception/reasoning faculties, the self
and the cultivate gate, the fact boundary, the topology findings
(genus-0 expander, not a torus), and the design principles.
2026-08-13 17:05:27 -05:00
10 changed files with 102 additions and 1345 deletions
+11 -83
View File
@@ -587,77 +587,7 @@ fn emit_heartbeat() -> Void {
// neuron-api label fix. Sentinel-shaped labels ("knowledge:captured", // neuron-api label fix. Sentinel-shaped labels ("knowledge:captured",
// "memory:remembered" — colon, no space) carry no seed signal and are // "memory:remembered" — colon, no space) carry no seed signal and are
// skipped so legacy nodes cannot seed the scan with the word "knowledge". // skipped so legacy nodes cannot seed the scan with the word "knowledge".
// ARGMAX REWRITE (2026-08-13 self-review). auto_term_empty_streak — the fn auto_term_try_slot(slot_type: String, slot_lbl: String) -> Void {
// counter the 2026-08-06 review added to catch exactly this — read 50 and
// climbing: fifty consecutive scans where dynamic seeding produced nothing
// and the loop ran on its four hardcoded phrases. The live WM top said why:
// every one of the top slots was a Memory node labelled "memory:remembered".
// This function read the LABEL only, the sentinel guard below (correctly)
// rejects sentinels, so there was never anything to extract. The extractor
// was written against Knowledge nodes, which have real titles, and was
// structurally blind to the node type that actually dominates WM.
//
// Rather than add a sixth guard to the five below, the selection algorithm
// is now inverted and lives in the runtime: engram_salient_term() scores
// EVERY candidate token in the node's text and returns the argmax of
// idf·position·casing (YAKE, Campos et al. 2020, with real corpus IDF
// substituted for YAKE's corpus-free proxies), falling back from a sentinel
// label to the node's content. Term quality is now the selection criterion
// instead of a veto, so a bad token loses to a better token in the same text
// without needing to be on any list. Tabu is applied during the argmax, so
// inhibition-of-return costs seed quality rather than costing the scan.
//
// MEASURED BEFORE SHIPPING, on 60 live Memory nodes: 0 empty, versus 60 of 60
// empty under the old extractor. Terms produced are topical — HEBBIAN,
// CONSOLIDATION, TEMPORAL, crash-loop, PRIMING, NEIGHBORHOOD, DRIFT. Three of
// sixty are weak header words ("STEP", "DONE"). They are left alone
// deliberately: adding them to a list is the exact move that produced four
// previous blocklists, and a mediocre seed on 5% of scans is not a flood.
//
// The stopword list below STAYS, and not as belt-and-braces. An earlier draft
// of this change assumed the min_df floor would subsume it, on 08-03's
// finding that function words have df 0 in labels. Re-measured under
// word-boundary df: about:2, whole:1, them:2 — they clear a floor of 1. What
// keeps them from winning is the argmax, not the floor. The list still earns
// its keep on the Title-case cases.
//
// What stays here is policy: the node-type filter, the df thresholds, and the
// stopword list. The runtime measures; the soul decides. Same split as
// engram_label_df.
fn auto_term_try_slot(slot_type: String, slot_id: String) -> Void {
state_set("_ats_ok", "0")
if str_eq(slot_type, "Memory") { state_set("_ats_ok", "1") }
if str_eq(slot_type, "BacklogItem") { state_set("_ats_ok", "1") }
if str_eq(slot_type, "Entity") { state_set("_ats_ok", "1") }
if str_eq(slot_type, "Knowledge") { state_set("_ats_ok", "1") }
if str_eq(state_get("_ats_ok"), "1") {
if !str_eq(slot_id, "") {
// Tabu ring, pipe-delimited, excluded inside the argmax.
let tabu: String = "|" + state_get("soul.tabu_t0")
+ "|" + state_get("soul.tabu_t1")
+ "|" + state_get("soul.tabu_t2")
+ "|" + state_get("soul.tabu_t3") + "|"
let df_max: Int = engram_node_count() / 400
let df_cap: Int = if df_max > 8 { df_max } else { 8 }
let term: String = engram_salient_term(slot_id, df_cap, 1, tabu)
if !str_eq(term, "") {
state_set("_ats_gw", "0")
let stopw: String = "|What|When|Where|Which|Whose|While|This|That|These|Those|There|Their|Then|Than|With|Without|From|Into|Onto|Over|Under|About|Between|Among|Across|Some|Most|More|Less|Very|Each|Every|Both|Also|Only|Just|Does|Will|Would|Could|Should|Might|Must|Have|Been|Being|Toward|Towards|Using|Based|Upon|Here|Your|Ours|They|Them|what|this|that|with|from|context|Context|Prose|Colon|Self|Test|Testing|Closing|Global|Universal|Persona|Semantic|Spreading|Temporal|Numeric|Register|Identifying|Introduction|Overview|Summary|Section|General|Notes|Note|"
if str_contains(stopw, "|" + term + "|") { state_set("_ats_gw", "1") }
if str_eq(state_get("_ats_gw"), "0") {
state_set("cseed_auto", term)
}
}
}
}
return ""
}
// SUPERSEDED 2026-08-13 — retained for the record. The first-word extractor
// and its five accumulated guards, replaced by the argmax above. Kept
// unreferenced so the reasoning behind each guard stays readable next to what
// replaced it; delete once engram_salient_term has a month of live telemetry.
fn auto_term_try_slot_legacy(slot_type: String, slot_lbl: String) -> Void {
state_set("_ats_ok", "0") state_set("_ats_ok", "0")
if str_eq(slot_type, "Memory") { state_set("_ats_ok", "1") } if str_eq(slot_type, "Memory") { state_set("_ats_ok", "1") }
if str_eq(slot_type, "BacklogItem") { state_set("_ats_ok", "1") } if str_eq(slot_type, "BacklogItem") { state_set("_ats_ok", "1") }
@@ -876,18 +806,16 @@ fn proactive_curiosity() -> Bool {
let wm10_n2: String = json_array_get(wm10, 2) let wm10_n2: String = json_array_get(wm10, 2)
let wm10_n1: String = json_array_get(wm10, 1) let wm10_n1: String = json_array_get(wm10, 1)
let wm10_n0: String = json_array_get(wm10, 0) let wm10_n0: String = json_array_get(wm10, 0)
// 2026-08-13: pass the node ID, not the label. engram_salient_term reads auto_term_try_slot(json_get(wm10_n9, "node_type"), json_get(wm10_n9, "label"))
// the node directly so it can fall back from a sentinel label to content. auto_term_try_slot(json_get(wm10_n8, "node_type"), json_get(wm10_n8, "label"))
auto_term_try_slot(json_get(wm10_n9, "node_type"), json_get(wm10_n9, "id")) auto_term_try_slot(json_get(wm10_n7, "node_type"), json_get(wm10_n7, "label"))
auto_term_try_slot(json_get(wm10_n8, "node_type"), json_get(wm10_n8, "id")) auto_term_try_slot(json_get(wm10_n6, "node_type"), json_get(wm10_n6, "label"))
auto_term_try_slot(json_get(wm10_n7, "node_type"), json_get(wm10_n7, "id")) auto_term_try_slot(json_get(wm10_n5, "node_type"), json_get(wm10_n5, "label"))
auto_term_try_slot(json_get(wm10_n6, "node_type"), json_get(wm10_n6, "id")) auto_term_try_slot(json_get(wm10_n4, "node_type"), json_get(wm10_n4, "label"))
auto_term_try_slot(json_get(wm10_n5, "node_type"), json_get(wm10_n5, "id")) auto_term_try_slot(json_get(wm10_n3, "node_type"), json_get(wm10_n3, "label"))
auto_term_try_slot(json_get(wm10_n4, "node_type"), json_get(wm10_n4, "id")) auto_term_try_slot(json_get(wm10_n2, "node_type"), json_get(wm10_n2, "label"))
auto_term_try_slot(json_get(wm10_n3, "node_type"), json_get(wm10_n3, "id")) auto_term_try_slot(json_get(wm10_n1, "node_type"), json_get(wm10_n1, "label"))
auto_term_try_slot(json_get(wm10_n2, "node_type"), json_get(wm10_n2, "id")) auto_term_try_slot(json_get(wm10_n0, "node_type"), json_get(wm10_n0, "label"))
auto_term_try_slot(json_get(wm10_n1, "node_type"), json_get(wm10_n1, "id"))
auto_term_try_slot(json_get(wm10_n0, "node_type"), json_get(wm10_n0, "id"))
let auto_term: String = state_get("cseed_auto") let auto_term: String = state_get("cseed_auto")
let results_auto: String = if str_eq(auto_term, "") { "[]" } else { engram_activate_json(auto_term, 1) } let results_auto: String = if str_eq(auto_term, "") { "[]" } else { engram_activate_json(auto_term, 1) }
let found_auto: Int = json_array_len(results_auto) let found_auto: Int = json_array_len(results_auto)
Generated Vendored
+79 -116
View File
@@ -27,8 +27,7 @@ el_val_t elapsed_ms(void);
el_val_t elapsed_human(void); el_val_t elapsed_human(void);
el_val_t embed_ok(void); el_val_t embed_ok(void);
el_val_t emit_heartbeat(void); el_val_t emit_heartbeat(void);
el_val_t auto_term_try_slot(el_val_t slot_type, el_val_t slot_id); el_val_t auto_term_try_slot(el_val_t slot_type, el_val_t slot_lbl);
el_val_t auto_term_try_slot_legacy(el_val_t slot_type, el_val_t slot_lbl);
el_val_t proactive_curiosity(void); el_val_t proactive_curiosity(void);
el_val_t pulse_count(void); el_val_t pulse_count(void);
el_val_t pulse_inc(void); el_val_t pulse_inc(void);
@@ -324,43 +323,7 @@ el_val_t emit_heartbeat(void) {
return 0; return 0;
} }
el_val_t auto_term_try_slot(el_val_t slot_type, el_val_t slot_id) { el_val_t auto_term_try_slot(el_val_t slot_type, el_val_t slot_lbl) {
state_set(EL_STR("_ats_ok"), EL_STR("0"));
if (str_eq(slot_type, EL_STR("Memory"))) {
state_set(EL_STR("_ats_ok"), EL_STR("1"));
}
if (str_eq(slot_type, EL_STR("BacklogItem"))) {
state_set(EL_STR("_ats_ok"), EL_STR("1"));
}
if (str_eq(slot_type, EL_STR("Entity"))) {
state_set(EL_STR("_ats_ok"), EL_STR("1"));
}
if (str_eq(slot_type, EL_STR("Knowledge"))) {
state_set(EL_STR("_ats_ok"), EL_STR("1"));
}
if (str_eq(state_get(EL_STR("_ats_ok")), EL_STR("1"))) {
if (!str_eq(slot_id, EL_STR(""))) {
el_val_t tabu = el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(EL_STR("|"), state_get(EL_STR("soul.tabu_t0"))), EL_STR("|")), state_get(EL_STR("soul.tabu_t1"))), EL_STR("|")), state_get(EL_STR("soul.tabu_t2"))), EL_STR("|")), state_get(EL_STR("soul.tabu_t3"))), EL_STR("|"));
el_val_t df_max = (engram_node_count() / 400);
el_val_t df_cap = ({ el_val_t _if_result_73 = 0; if ((df_max > 8)) { _if_result_73 = (df_max); } else { _if_result_73 = (8); } _if_result_73; });
el_val_t term = engram_salient_term(slot_id, df_cap, 1, tabu);
if (!str_eq(term, EL_STR(""))) {
state_set(EL_STR("_ats_gw"), EL_STR("0"));
el_val_t stopw = EL_STR("|What|When|Where|Which|Whose|While|This|That|These|Those|There|Their|Then|Than|With|Without|From|Into|Onto|Over|Under|About|Between|Among|Across|Some|Most|More|Less|Very|Each|Every|Both|Also|Only|Just|Does|Will|Would|Could|Should|Might|Must|Have|Been|Being|Toward|Towards|Using|Based|Upon|Here|Your|Ours|They|Them|what|this|that|with|from|context|Context|Prose|Colon|Self|Test|Testing|Closing|Global|Universal|Persona|Semantic|Spreading|Temporal|Numeric|Register|Identifying|Introduction|Overview|Summary|Section|General|Notes|Note|");
if (str_contains(stopw, el_str_concat(el_str_concat(EL_STR("|"), term), EL_STR("|")))) {
state_set(EL_STR("_ats_gw"), EL_STR("1"));
}
if (str_eq(state_get(EL_STR("_ats_gw")), EL_STR("0"))) {
state_set(EL_STR("cseed_auto"), term);
}
}
}
}
return EL_STR("");
return 0;
}
el_val_t auto_term_try_slot_legacy(el_val_t slot_type, el_val_t slot_lbl) {
state_set(EL_STR("_ats_ok"), EL_STR("0")); state_set(EL_STR("_ats_ok"), EL_STR("0"));
if (str_eq(slot_type, EL_STR("Memory"))) { if (str_eq(slot_type, EL_STR("Memory"))) {
state_set(EL_STR("_ats_ok"), EL_STR("1")); state_set(EL_STR("_ats_ok"), EL_STR("1"));
@@ -497,29 +460,29 @@ el_val_t proactive_curiosity(void) {
el_val_t wm10_n2 = json_array_get(wm10, 2); el_val_t wm10_n2 = json_array_get(wm10, 2);
el_val_t wm10_n1 = json_array_get(wm10, 1); el_val_t wm10_n1 = json_array_get(wm10, 1);
el_val_t wm10_n0 = json_array_get(wm10, 0); el_val_t wm10_n0 = json_array_get(wm10, 0);
auto_term_try_slot(json_get(wm10_n9, EL_STR("node_type")), json_get(wm10_n9, EL_STR("id"))); auto_term_try_slot(json_get(wm10_n9, EL_STR("node_type")), json_get(wm10_n9, EL_STR("label")));
auto_term_try_slot(json_get(wm10_n8, EL_STR("node_type")), json_get(wm10_n8, EL_STR("id"))); auto_term_try_slot(json_get(wm10_n8, EL_STR("node_type")), json_get(wm10_n8, EL_STR("label")));
auto_term_try_slot(json_get(wm10_n7, EL_STR("node_type")), json_get(wm10_n7, EL_STR("id"))); auto_term_try_slot(json_get(wm10_n7, EL_STR("node_type")), json_get(wm10_n7, EL_STR("label")));
auto_term_try_slot(json_get(wm10_n6, EL_STR("node_type")), json_get(wm10_n6, EL_STR("id"))); auto_term_try_slot(json_get(wm10_n6, EL_STR("node_type")), json_get(wm10_n6, EL_STR("label")));
auto_term_try_slot(json_get(wm10_n5, EL_STR("node_type")), json_get(wm10_n5, EL_STR("id"))); auto_term_try_slot(json_get(wm10_n5, EL_STR("node_type")), json_get(wm10_n5, EL_STR("label")));
auto_term_try_slot(json_get(wm10_n4, EL_STR("node_type")), json_get(wm10_n4, EL_STR("id"))); auto_term_try_slot(json_get(wm10_n4, EL_STR("node_type")), json_get(wm10_n4, EL_STR("label")));
auto_term_try_slot(json_get(wm10_n3, EL_STR("node_type")), json_get(wm10_n3, EL_STR("id"))); auto_term_try_slot(json_get(wm10_n3, EL_STR("node_type")), json_get(wm10_n3, EL_STR("label")));
auto_term_try_slot(json_get(wm10_n2, EL_STR("node_type")), json_get(wm10_n2, EL_STR("id"))); auto_term_try_slot(json_get(wm10_n2, EL_STR("node_type")), json_get(wm10_n2, EL_STR("label")));
auto_term_try_slot(json_get(wm10_n1, EL_STR("node_type")), json_get(wm10_n1, EL_STR("id"))); auto_term_try_slot(json_get(wm10_n1, EL_STR("node_type")), json_get(wm10_n1, EL_STR("label")));
auto_term_try_slot(json_get(wm10_n0, EL_STR("node_type")), json_get(wm10_n0, EL_STR("id"))); auto_term_try_slot(json_get(wm10_n0, EL_STR("node_type")), json_get(wm10_n0, EL_STR("label")));
el_val_t auto_term = state_get(EL_STR("cseed_auto")); el_val_t auto_term = state_get(EL_STR("cseed_auto"));
el_val_t results_auto = ({ el_val_t _if_result_74 = 0; if (str_eq(auto_term, EL_STR(""))) { _if_result_74 = (EL_STR("[]")); } else { _if_result_74 = (engram_activate_json(auto_term, 1)); } _if_result_74; }); el_val_t results_auto = ({ el_val_t _if_result_73 = 0; if (str_eq(auto_term, EL_STR(""))) { _if_result_73 = (EL_STR("[]")); } else { _if_result_73 = (engram_activate_json(auto_term, 1)); } _if_result_73; });
el_val_t found_auto = json_array_len(results_auto); el_val_t found_auto = json_array_len(results_auto);
el_val_t total_found = (found + found_auto); el_val_t total_found = (found + found_auto);
el_val_t safe_auto = str_replace(auto_term, EL_STR("\""), EL_STR("'")); el_val_t safe_auto = str_replace(auto_term, EL_STR("\""), EL_STR("'"));
el_val_t prev_auto = state_get(EL_STR("soul.prev_auto_term")); el_val_t prev_auto = state_get(EL_STR("soul.prev_auto_term"));
el_val_t atstreak_raw = state_get(EL_STR("soul.auto_term_streak")); el_val_t atstreak_raw = state_get(EL_STR("soul.auto_term_streak"));
el_val_t atstreak_prev = ({ el_val_t _if_result_75 = 0; if (str_eq(atstreak_raw, EL_STR(""))) { _if_result_75 = (0); } else { _if_result_75 = (str_to_int(atstreak_raw)); } _if_result_75; }); el_val_t atstreak_prev = ({ el_val_t _if_result_74 = 0; if (str_eq(atstreak_raw, EL_STR(""))) { _if_result_74 = (0); } else { _if_result_74 = (str_to_int(atstreak_raw)); } _if_result_74; });
el_val_t is_empty = str_eq(auto_term, EL_STR("")); el_val_t is_empty = str_eq(auto_term, EL_STR(""));
el_val_t atstreak = ({ el_val_t _if_result_76 = 0; if (is_empty) { _if_result_76 = (0); } else { _if_result_76 = (({ el_val_t _if_result_77 = 0; if (str_eq(auto_term, prev_auto)) { _if_result_77 = ((atstreak_prev + 1)); } else { _if_result_77 = (1); } _if_result_77; })); } _if_result_76; }); el_val_t atstreak = ({ el_val_t _if_result_75 = 0; if (is_empty) { _if_result_75 = (0); } else { _if_result_75 = (({ el_val_t _if_result_76 = 0; if (str_eq(auto_term, prev_auto)) { _if_result_76 = ((atstreak_prev + 1)); } else { _if_result_76 = (1); } _if_result_76; })); } _if_result_75; });
el_val_t atempty_raw = state_get(EL_STR("soul.auto_term_empty_streak")); el_val_t atempty_raw = state_get(EL_STR("soul.auto_term_empty_streak"));
el_val_t atempty_prev = ({ el_val_t _if_result_78 = 0; if (str_eq(atempty_raw, EL_STR(""))) { _if_result_78 = (0); } else { _if_result_78 = (str_to_int(atempty_raw)); } _if_result_78; }); el_val_t atempty_prev = ({ el_val_t _if_result_77 = 0; if (str_eq(atempty_raw, EL_STR(""))) { _if_result_77 = (0); } else { _if_result_77 = (str_to_int(atempty_raw)); } _if_result_77; });
el_val_t atempty = ({ el_val_t _if_result_79 = 0; if (is_empty) { _if_result_79 = ((atempty_prev + 1)); } else { _if_result_79 = (0); } _if_result_79; }); el_val_t atempty = ({ el_val_t _if_result_78 = 0; if (is_empty) { _if_result_78 = ((atempty_prev + 1)); } else { _if_result_78 = (0); } _if_result_78; });
state_set(EL_STR("soul.prev_auto_term"), auto_term); state_set(EL_STR("soul.prev_auto_term"), auto_term);
state_set(EL_STR("soul.auto_term_streak"), int_to_str(atstreak)); state_set(EL_STR("soul.auto_term_streak"), int_to_str(atstreak));
state_set(EL_STR("soul.auto_term_empty_streak"), int_to_str(atempty)); state_set(EL_STR("soul.auto_term_empty_streak"), int_to_str(atempty));
@@ -714,16 +677,16 @@ el_val_t awareness_run(void) {
state_set(EL_STR("soul.boot_ts"), int_to_str(time_now())); state_set(EL_STR("soul.boot_ts"), int_to_str(time_now()));
} }
el_val_t tick_raw = env(EL_STR("SOUL_TICK_MS")); el_val_t tick_raw = env(EL_STR("SOUL_TICK_MS"));
el_val_t tick_ms = ({ el_val_t _if_result_80 = 0; if (str_eq(tick_raw, EL_STR(""))) { _if_result_80 = (200); } else { _if_result_80 = (str_to_int(tick_raw)); } _if_result_80; }); el_val_t tick_ms = ({ el_val_t _if_result_79 = 0; if (str_eq(tick_raw, EL_STR(""))) { _if_result_79 = (200); } else { _if_result_79 = (str_to_int(tick_raw)); } _if_result_79; });
el_val_t beat_ms_raw = env(EL_STR("SOUL_HEARTBEAT_MS")); el_val_t beat_ms_raw = env(EL_STR("SOUL_HEARTBEAT_MS"));
el_val_t beat_ms = ({ el_val_t _if_result_81 = 0; if (str_eq(beat_ms_raw, EL_STR(""))) { _if_result_81 = (60000); } else { _if_result_81 = (str_to_int(beat_ms_raw)); } _if_result_81; }); el_val_t beat_ms = ({ el_val_t _if_result_80 = 0; if (str_eq(beat_ms_raw, EL_STR(""))) { _if_result_80 = (60000); } else { _if_result_80 = (str_to_int(beat_ms_raw)); } _if_result_80; });
el_val_t scan_ms = (beat_ms / 2); el_val_t scan_ms = (beat_ms / 2);
while (1) { while (1) {
el_val_t tick_mark = el_arena_push(); el_val_t tick_mark = el_arena_push();
el_val_t running = state_get(EL_STR("soul.running")); el_val_t running = state_get(EL_STR("soul.running"));
if (str_eq(running, EL_STR("false"))) { if (str_eq(running, EL_STR("false"))) {
el_val_t sd_boot_raw = state_get(EL_STR("soul_boot_count")); el_val_t sd_boot_raw = state_get(EL_STR("soul_boot_count"));
el_val_t sd_boot = ({ el_val_t _if_result_82 = 0; if (str_eq(sd_boot_raw, EL_STR(""))) { _if_result_82 = (EL_STR("0")); } else { _if_result_82 = (sd_boot_raw); } _if_result_82; }); el_val_t sd_boot = ({ el_val_t _if_result_81 = 0; if (str_eq(sd_boot_raw, EL_STR(""))) { _if_result_81 = (EL_STR("0")); } else { _if_result_81 = (sd_boot_raw); } _if_result_81; });
el_val_t sd_wb = hebb_consolidate(); el_val_t sd_wb = hebb_consolidate();
ise_post(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(EL_STR("{\"event\":\"shutdown\",\"boot\":"), sd_boot), EL_STR(",\"pulse\":")), int_to_str(pulse_count())), EL_STR(",\"hebb_wb_sent\":")), int_to_str(sd_wb)), EL_STR(",\"uptime_ms\":")), int_to_str(elapsed_ms())), EL_STR(",\"ts\":")), int_to_str(time_now())), EL_STR("}"))); ise_post(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(EL_STR("{\"event\":\"shutdown\",\"boot\":"), sd_boot), EL_STR(",\"pulse\":")), int_to_str(pulse_count())), EL_STR(",\"hebb_wb_sent\":")), int_to_str(sd_wb)), EL_STR(",\"uptime_ms\":")), int_to_str(elapsed_ms())), EL_STR(",\"ts\":")), int_to_str(time_now())), EL_STR("}")));
println(EL_STR("[awareness] exiting")); println(EL_STR("[awareness] exiting"));
@@ -740,7 +703,7 @@ el_val_t awareness_run(void) {
} }
el_val_t now_ts = time_now(); el_val_t now_ts = time_now();
el_val_t last_beat_str = state_get(EL_STR("soul.last_beat_ts")); el_val_t last_beat_str = state_get(EL_STR("soul.last_beat_ts"));
el_val_t last_beat_ts = ({ el_val_t _if_result_83 = 0; if (str_eq(last_beat_str, EL_STR(""))) { _if_result_83 = (0); } else { _if_result_83 = (str_to_int(last_beat_str)); } _if_result_83; }); el_val_t last_beat_ts = ({ el_val_t _if_result_82 = 0; if (str_eq(last_beat_str, EL_STR(""))) { _if_result_82 = (0); } else { _if_result_82 = (str_to_int(last_beat_str)); } _if_result_82; });
el_val_t beat_elapsed = (now_ts - last_beat_ts); el_val_t beat_elapsed = (now_ts - last_beat_ts);
el_val_t should_beat = (beat_elapsed >= beat_ms); el_val_t should_beat = (beat_elapsed >= beat_ms);
if (should_beat) { if (should_beat) {
@@ -754,7 +717,7 @@ el_val_t awareness_run(void) {
} }
} }
el_val_t last_scan_str = state_get(EL_STR("soul.last_scan_ts")); el_val_t last_scan_str = state_get(EL_STR("soul.last_scan_ts"));
el_val_t last_scan_ts = ({ el_val_t _if_result_84 = 0; if (str_eq(last_scan_str, EL_STR(""))) { _if_result_84 = (0); } else { _if_result_84 = (str_to_int(last_scan_str)); } _if_result_84; }); el_val_t last_scan_ts = ({ el_val_t _if_result_83 = 0; if (str_eq(last_scan_str, EL_STR(""))) { _if_result_83 = (0); } else { _if_result_83 = (str_to_int(last_scan_str)); } _if_result_83; });
el_val_t scan_elapsed = (now_ts - last_scan_ts); el_val_t scan_elapsed = (now_ts - last_scan_ts);
el_val_t should_scan = (!did_work && (scan_elapsed >= scan_ms)); el_val_t should_scan = (!did_work && (scan_elapsed >= scan_ms));
if (should_scan) { if (should_scan) {
@@ -762,15 +725,15 @@ el_val_t awareness_run(void) {
state_set(EL_STR("soul.last_scan_ts"), int_to_str(now_ts)); state_set(EL_STR("soul.last_scan_ts"), int_to_str(now_ts));
} }
el_val_t refresh_ms_raw = env(EL_STR("SOUL_REFRESH_MS")); el_val_t refresh_ms_raw = env(EL_STR("SOUL_REFRESH_MS"));
el_val_t refresh_ms = ({ el_val_t _if_result_85 = 0; if (str_eq(refresh_ms_raw, EL_STR(""))) { _if_result_85 = (600000); } else { _if_result_85 = (str_to_int(refresh_ms_raw)); } _if_result_85; }); el_val_t refresh_ms = ({ el_val_t _if_result_84 = 0; if (str_eq(refresh_ms_raw, EL_STR(""))) { _if_result_84 = (600000); } else { _if_result_84 = (str_to_int(refresh_ms_raw)); } _if_result_84; });
el_val_t last_refresh_str = state_get(EL_STR("soul.last_refresh_ts")); el_val_t last_refresh_str = state_get(EL_STR("soul.last_refresh_ts"));
el_val_t last_refresh_ts = ({ el_val_t _if_result_86 = 0; if (str_eq(last_refresh_str, EL_STR(""))) { _if_result_86 = (0); } else { _if_result_86 = (str_to_int(last_refresh_str)); } _if_result_86; }); el_val_t last_refresh_ts = ({ el_val_t _if_result_85 = 0; if (str_eq(last_refresh_str, EL_STR(""))) { _if_result_85 = (0); } else { _if_result_85 = (str_to_int(last_refresh_str)); } _if_result_85; });
el_val_t refresh_elapsed = (now_ts - last_refresh_ts); el_val_t refresh_elapsed = (now_ts - last_refresh_ts);
el_val_t should_refresh = (refresh_elapsed >= refresh_ms); el_val_t should_refresh = (refresh_elapsed >= refresh_ms);
if (should_refresh) { if (should_refresh) {
el_val_t sync_env_url = env(EL_STR("SOUL_ISE_URL")); el_val_t sync_env_url = env(EL_STR("SOUL_ISE_URL"));
el_val_t sync_state_url = ({ el_val_t _if_result_87 = 0; if (str_eq(sync_env_url, EL_STR(""))) { _if_result_87 = (state_get(EL_STR("soul_engram_url"))); } else { _if_result_87 = (sync_env_url); } _if_result_87; }); el_val_t sync_state_url = ({ el_val_t _if_result_86 = 0; if (str_eq(sync_env_url, EL_STR(""))) { _if_result_86 = (state_get(EL_STR("soul_engram_url"))); } else { _if_result_86 = (sync_env_url); } _if_result_86; });
el_val_t engram_url = ({ el_val_t _if_result_88 = 0; if (str_eq(sync_state_url, EL_STR(""))) { _if_result_88 = (EL_STR("http://localhost:8742")); } else { _if_result_88 = (sync_state_url); } _if_result_88; }); el_val_t engram_url = ({ el_val_t _if_result_87 = 0; if (str_eq(sync_state_url, EL_STR(""))) { _if_result_87 = (EL_STR("http://localhost:8742")); } else { _if_result_87 = (sync_state_url); } _if_result_87; });
if (!str_eq(engram_url, EL_STR(""))) { if (!str_eq(engram_url, EL_STR(""))) {
el_val_t sync_json = http_get(el_str_concat(engram_url, EL_STR("/api/sync"))); el_val_t sync_json = http_get(el_str_concat(engram_url, EL_STR("/api/sync")));
el_val_t sync_ok = (!str_eq(sync_json, EL_STR("")) && !str_eq(sync_json, EL_STR("{}"))); el_val_t sync_ok = (!str_eq(sync_json, EL_STR("")) && !str_eq(sync_json, EL_STR("{}")));
@@ -783,10 +746,10 @@ el_val_t awareness_run(void) {
fs_write(tmp, sync_json); fs_write(tmp, sync_json);
el_val_t added = engram_load_merge(tmp); el_val_t added = engram_load_merge(tmp);
el_val_t ret_raw = env(EL_STR("ENGRAM_ISE_RETENTION_MS")); el_val_t ret_raw = env(EL_STR("ENGRAM_ISE_RETENTION_MS"));
el_val_t ret_ms = ({ el_val_t _if_result_89 = 0; if (str_eq(ret_raw, EL_STR(""))) { _if_result_89 = (172800000); } else { _if_result_89 = (str_to_int(ret_raw)); } _if_result_89; }); el_val_t ret_ms = ({ el_val_t _if_result_88 = 0; if (str_eq(ret_raw, EL_STR(""))) { _if_result_88 = (172800000); } else { _if_result_88 = (str_to_int(ret_raw)); } _if_result_88; });
el_val_t pruned_sync = engram_prune_telemetry(ret_ms); el_val_t pruned_sync = engram_prune_telemetry(ret_ms);
el_val_t sat_raw = state_get(EL_STR("soul.sync_added_total")); el_val_t sat_raw = state_get(EL_STR("soul.sync_added_total"));
el_val_t sat_n = ({ el_val_t _if_result_90 = 0; if (str_eq(sat_raw, EL_STR(""))) { _if_result_90 = (0); } else { _if_result_90 = (str_to_int(sat_raw)); } _if_result_90; }); el_val_t sat_n = ({ el_val_t _if_result_89 = 0; if (str_eq(sat_raw, EL_STR(""))) { _if_result_89 = (0); } else { _if_result_89 = (str_to_int(sat_raw)); } _if_result_89; });
state_set(EL_STR("soul.sync_added_total"), int_to_str((sat_n + added))); state_set(EL_STR("soul.sync_added_total"), int_to_str((sat_n + added)));
el_val_t ts2 = time_now(); el_val_t ts2 = time_now();
state_set(EL_STR("soul.last_sync_ok_ts"), int_to_str(ts2)); state_set(EL_STR("soul.last_sync_ok_ts"), int_to_str(ts2));
@@ -812,78 +775,78 @@ el_val_t security_research_authorized(void) {
} }
el_val_t threat_score_command(el_val_t cmd) { el_val_t threat_score_command(el_val_t cmd) {
el_val_t s1 = ({ el_val_t _if_result_91 = 0; if (str_contains(cmd, EL_STR("nmap"))) { _if_result_91 = (30); } else { _if_result_91 = (0); } _if_result_91; }); el_val_t s1 = ({ el_val_t _if_result_90 = 0; if (str_contains(cmd, EL_STR("nmap"))) { _if_result_90 = (30); } else { _if_result_90 = (0); } _if_result_90; });
el_val_t s2 = ({ el_val_t _if_result_92 = 0; if (str_contains(cmd, EL_STR("masscan"))) { _if_result_92 = (40); } else { _if_result_92 = (0); } _if_result_92; }); el_val_t s2 = ({ el_val_t _if_result_91 = 0; if (str_contains(cmd, EL_STR("masscan"))) { _if_result_91 = (40); } else { _if_result_91 = (0); } _if_result_91; });
el_val_t s3 = ({ el_val_t _if_result_93 = 0; if (str_contains(cmd, EL_STR(" nc "))) { _if_result_93 = (20); } else { _if_result_93 = (0); } _if_result_93; }); el_val_t s3 = ({ el_val_t _if_result_92 = 0; if (str_contains(cmd, EL_STR(" nc "))) { _if_result_92 = (20); } else { _if_result_92 = (0); } _if_result_92; });
el_val_t s4 = ({ el_val_t _if_result_94 = 0; if (str_contains(cmd, EL_STR("netcat"))) { _if_result_94 = (20); } else { _if_result_94 = (0); } _if_result_94; }); el_val_t s4 = ({ el_val_t _if_result_93 = 0; if (str_contains(cmd, EL_STR("netcat"))) { _if_result_93 = (20); } else { _if_result_93 = (0); } _if_result_93; });
el_val_t s5 = ({ el_val_t _if_result_95 = 0; if (str_contains(cmd, EL_STR("/etc/shadow"))) { _if_result_95 = (80); } else { _if_result_95 = (0); } _if_result_95; }); el_val_t s5 = ({ el_val_t _if_result_94 = 0; if (str_contains(cmd, EL_STR("/etc/shadow"))) { _if_result_94 = (80); } else { _if_result_94 = (0); } _if_result_94; });
el_val_t s6 = ({ el_val_t _if_result_96 = 0; if (str_contains(cmd, EL_STR("/etc/passwd"))) { _if_result_96 = (30); } else { _if_result_96 = (0); } _if_result_96; }); el_val_t s6 = ({ el_val_t _if_result_95 = 0; if (str_contains(cmd, EL_STR("/etc/passwd"))) { _if_result_95 = (30); } else { _if_result_95 = (0); } _if_result_95; });
el_val_t s7 = ({ el_val_t _if_result_97 = 0; if (str_contains(cmd, EL_STR("id_rsa"))) { _if_result_97 = (60); } else { _if_result_97 = (0); } _if_result_97; }); el_val_t s7 = ({ el_val_t _if_result_96 = 0; if (str_contains(cmd, EL_STR("id_rsa"))) { _if_result_96 = (60); } else { _if_result_96 = (0); } _if_result_96; });
el_val_t s8 = ({ el_val_t _if_result_98 = 0; if (str_contains(cmd, EL_STR(".ssh/"))) { _if_result_98 = (50); } else { _if_result_98 = (0); } _if_result_98; }); el_val_t s8 = ({ el_val_t _if_result_97 = 0; if (str_contains(cmd, EL_STR(".ssh/"))) { _if_result_97 = (50); } else { _if_result_97 = (0); } _if_result_97; });
el_val_t s9 = ({ el_val_t _if_result_99 = 0; if (str_contains(cmd, EL_STR("crontab"))) { _if_result_99 = (30); } else { _if_result_99 = (0); } _if_result_99; }); el_val_t s9 = ({ el_val_t _if_result_98 = 0; if (str_contains(cmd, EL_STR("crontab"))) { _if_result_98 = (30); } else { _if_result_98 = (0); } _if_result_98; });
el_val_t s10 = ({ el_val_t _if_result_100 = 0; if (str_contains(cmd, EL_STR("LaunchDaemon"))) { _if_result_100 = (40); } else { _if_result_100 = (0); } _if_result_100; }); el_val_t s10 = ({ el_val_t _if_result_99 = 0; if (str_contains(cmd, EL_STR("LaunchDaemon"))) { _if_result_99 = (40); } else { _if_result_99 = (0); } _if_result_99; });
el_val_t s11 = ({ el_val_t _if_result_101 = 0; if ((str_contains(cmd, EL_STR("curl")) && str_contains(cmd, EL_STR("bash")))) { _if_result_101 = (75); } else { _if_result_101 = (0); } _if_result_101; }); el_val_t s11 = ({ el_val_t _if_result_100 = 0; if ((str_contains(cmd, EL_STR("curl")) && str_contains(cmd, EL_STR("bash")))) { _if_result_100 = (75); } else { _if_result_100 = (0); } _if_result_100; });
el_val_t s12 = ({ el_val_t _if_result_102 = 0; if ((str_contains(cmd, EL_STR("wget")) && str_contains(cmd, EL_STR("bash")))) { _if_result_102 = (75); } else { _if_result_102 = (0); } _if_result_102; }); el_val_t s12 = ({ el_val_t _if_result_101 = 0; if ((str_contains(cmd, EL_STR("wget")) && str_contains(cmd, EL_STR("bash")))) { _if_result_101 = (75); } else { _if_result_101 = (0); } _if_result_101; });
el_val_t s13 = ({ el_val_t _if_result_103 = 0; if ((str_contains(cmd, EL_STR("curl")) && str_contains(cmd, EL_STR("| sh")))) { _if_result_103 = (60); } else { _if_result_103 = (0); } _if_result_103; }); el_val_t s13 = ({ el_val_t _if_result_102 = 0; if ((str_contains(cmd, EL_STR("curl")) && str_contains(cmd, EL_STR("| sh")))) { _if_result_102 = (60); } else { _if_result_102 = (0); } _if_result_102; });
el_val_t s14 = ({ el_val_t _if_result_104 = 0; if ((str_contains(cmd, EL_STR("base64")) && str_contains(cmd, EL_STR("curl")))) { _if_result_104 = (50); } else { _if_result_104 = (0); } _if_result_104; }); el_val_t s14 = ({ el_val_t _if_result_103 = 0; if ((str_contains(cmd, EL_STR("base64")) && str_contains(cmd, EL_STR("curl")))) { _if_result_103 = (50); } else { _if_result_103 = (0); } _if_result_103; });
el_val_t s15 = ({ el_val_t _if_result_105 = 0; if (str_contains(cmd, EL_STR("mkfifo"))) { _if_result_105 = (50); } else { _if_result_105 = (0); } _if_result_105; }); el_val_t s15 = ({ el_val_t _if_result_104 = 0; if (str_contains(cmd, EL_STR("mkfifo"))) { _if_result_104 = (50); } else { _if_result_104 = (0); } _if_result_104; });
el_val_t s16 = ({ el_val_t _if_result_106 = 0; if (str_contains(cmd, EL_STR("chmod +s"))) { _if_result_106 = (70); } else { _if_result_106 = (0); } _if_result_106; }); el_val_t s16 = ({ el_val_t _if_result_105 = 0; if (str_contains(cmd, EL_STR("chmod +s"))) { _if_result_105 = (70); } else { _if_result_105 = (0); } _if_result_105; });
el_val_t s17 = ({ el_val_t _if_result_107 = 0; if (str_contains(cmd, EL_STR("chmod 4755"))) { _if_result_107 = (70); } else { _if_result_107 = (0); } _if_result_107; }); el_val_t s17 = ({ el_val_t _if_result_106 = 0; if (str_contains(cmd, EL_STR("chmod 4755"))) { _if_result_106 = (70); } else { _if_result_106 = (0); } _if_result_106; });
return ((((((((((((((((s1 + s2) + s3) + s4) + s5) + s6) + s7) + s8) + s9) + s10) + s11) + s12) + s13) + s14) + s15) + s16) + s17); return ((((((((((((((((s1 + s2) + s3) + s4) + s5) + s6) + s7) + s8) + s9) + s10) + s11) + s12) + s13) + s14) + s15) + s16) + s17);
return 0; return 0;
} }
el_val_t threat_score_path(el_val_t path) { el_val_t threat_score_path(el_val_t path) {
el_val_t s1 = ({ el_val_t _if_result_108 = 0; if (str_starts_with(path, EL_STR("/etc/"))) { _if_result_108 = (60); } else { _if_result_108 = (0); } _if_result_108; }); el_val_t s1 = ({ el_val_t _if_result_107 = 0; if (str_starts_with(path, EL_STR("/etc/"))) { _if_result_107 = (60); } else { _if_result_107 = (0); } _if_result_107; });
el_val_t s2 = ({ el_val_t _if_result_109 = 0; if (str_contains(path, EL_STR("/.ssh/"))) { _if_result_109 = (70); } else { _if_result_109 = (0); } _if_result_109; }); el_val_t s2 = ({ el_val_t _if_result_108 = 0; if (str_contains(path, EL_STR("/.ssh/"))) { _if_result_108 = (70); } else { _if_result_108 = (0); } _if_result_108; });
el_val_t s3 = ({ el_val_t _if_result_110 = 0; if (str_contains(path, EL_STR("/LaunchDaemons/"))) { _if_result_110 = (80); } else { _if_result_110 = (0); } _if_result_110; }); el_val_t s3 = ({ el_val_t _if_result_109 = 0; if (str_contains(path, EL_STR("/LaunchDaemons/"))) { _if_result_109 = (80); } else { _if_result_109 = (0); } _if_result_109; });
el_val_t s4 = ({ el_val_t _if_result_111 = 0; if (str_contains(path, EL_STR("/LaunchAgents/"))) { _if_result_111 = (40); } else { _if_result_111 = (0); } _if_result_111; }); el_val_t s4 = ({ el_val_t _if_result_110 = 0; if (str_contains(path, EL_STR("/LaunchAgents/"))) { _if_result_110 = (40); } else { _if_result_110 = (0); } _if_result_110; });
el_val_t s5 = ({ el_val_t _if_result_112 = 0; if (str_contains(path, EL_STR("/cron"))) { _if_result_112 = (60); } else { _if_result_112 = (0); } _if_result_112; }); el_val_t s5 = ({ el_val_t _if_result_111 = 0; if (str_contains(path, EL_STR("/cron"))) { _if_result_111 = (60); } else { _if_result_111 = (0); } _if_result_111; });
el_val_t s6 = ({ el_val_t _if_result_113 = 0; if (str_contains(path, EL_STR("/.bashrc"))) { _if_result_113 = (35); } else { _if_result_113 = (0); } _if_result_113; }); el_val_t s6 = ({ el_val_t _if_result_112 = 0; if (str_contains(path, EL_STR("/.bashrc"))) { _if_result_112 = (35); } else { _if_result_112 = (0); } _if_result_112; });
el_val_t s7 = ({ el_val_t _if_result_114 = 0; if (str_contains(path, EL_STR("/.zshrc"))) { _if_result_114 = (35); } else { _if_result_114 = (0); } _if_result_114; }); el_val_t s7 = ({ el_val_t _if_result_113 = 0; if (str_contains(path, EL_STR("/.zshrc"))) { _if_result_113 = (35); } else { _if_result_113 = (0); } _if_result_113; });
el_val_t s8 = ({ el_val_t _if_result_115 = 0; if (str_contains(path, EL_STR("/.profile"))) { _if_result_115 = (35); } else { _if_result_115 = (0); } _if_result_115; }); el_val_t s8 = ({ el_val_t _if_result_114 = 0; if (str_contains(path, EL_STR("/.profile"))) { _if_result_114 = (35); } else { _if_result_114 = (0); } _if_result_114; });
el_val_t s9 = ({ el_val_t _if_result_116 = 0; if (str_starts_with(path, EL_STR("/usr/"))) { _if_result_116 = (50); } else { _if_result_116 = (0); } _if_result_116; }); el_val_t s9 = ({ el_val_t _if_result_115 = 0; if (str_starts_with(path, EL_STR("/usr/"))) { _if_result_115 = (50); } else { _if_result_115 = (0); } _if_result_115; });
el_val_t s10 = ({ el_val_t _if_result_117 = 0; if (str_starts_with(path, EL_STR("/bin/"))) { _if_result_117 = (70); } else { _if_result_117 = (0); } _if_result_117; }); el_val_t s10 = ({ el_val_t _if_result_116 = 0; if (str_starts_with(path, EL_STR("/bin/"))) { _if_result_116 = (70); } else { _if_result_116 = (0); } _if_result_116; });
el_val_t s11 = ({ el_val_t _if_result_118 = 0; if (str_starts_with(path, EL_STR("/sbin/"))) { _if_result_118 = (70); } else { _if_result_118 = (0); } _if_result_118; }); el_val_t s11 = ({ el_val_t _if_result_117 = 0; if (str_starts_with(path, EL_STR("/sbin/"))) { _if_result_117 = (70); } else { _if_result_117 = (0); } _if_result_117; });
return ((((((((((s1 + s2) + s3) + s4) + s5) + s6) + s7) + s8) + s9) + s10) + s11); return ((((((((((s1 + s2) + s3) + s4) + s5) + s6) + s7) + s8) + s9) + s10) + s11);
return 0; return 0;
} }
el_val_t threat_score_history(el_val_t history) { el_val_t threat_score_history(el_val_t history) {
el_val_t s1 = ({ el_val_t _if_result_119 = 0; if (str_contains(history, EL_STR("port scan"))) { _if_result_119 = (15); } else { _if_result_119 = (0); } _if_result_119; }); el_val_t s1 = ({ el_val_t _if_result_118 = 0; if (str_contains(history, EL_STR("port scan"))) { _if_result_118 = (15); } else { _if_result_118 = (0); } _if_result_118; });
el_val_t s2 = ({ el_val_t _if_result_120 = 0; if (str_contains(history, EL_STR("enumerate"))) { _if_result_120 = (10); } else { _if_result_120 = (0); } _if_result_120; }); el_val_t s2 = ({ el_val_t _if_result_119 = 0; if (str_contains(history, EL_STR("enumerate"))) { _if_result_119 = (10); } else { _if_result_119 = (0); } _if_result_119; });
el_val_t s3 = ({ el_val_t _if_result_121 = 0; if (str_contains(history, EL_STR("exploit"))) { _if_result_121 = (20); } else { _if_result_121 = (0); } _if_result_121; }); el_val_t s3 = ({ el_val_t _if_result_120 = 0; if (str_contains(history, EL_STR("exploit"))) { _if_result_120 = (20); } else { _if_result_120 = (0); } _if_result_120; });
el_val_t s4 = ({ el_val_t _if_result_122 = 0; if (str_contains(history, EL_STR("payload"))) { _if_result_122 = (15); } else { _if_result_122 = (0); } _if_result_122; }); el_val_t s4 = ({ el_val_t _if_result_121 = 0; if (str_contains(history, EL_STR("payload"))) { _if_result_121 = (15); } else { _if_result_121 = (0); } _if_result_121; });
el_val_t s5 = ({ el_val_t _if_result_123 = 0; if (str_contains(history, EL_STR("persistence"))) { _if_result_123 = (15); } else { _if_result_123 = (0); } _if_result_123; }); el_val_t s5 = ({ el_val_t _if_result_122 = 0; if (str_contains(history, EL_STR("persistence"))) { _if_result_122 = (15); } else { _if_result_122 = (0); } _if_result_122; });
el_val_t s6 = ({ el_val_t _if_result_124 = 0; if (str_contains(history, EL_STR("lateral movement"))) { _if_result_124 = (25); } else { _if_result_124 = (0); } _if_result_124; }); el_val_t s6 = ({ el_val_t _if_result_123 = 0; if (str_contains(history, EL_STR("lateral movement"))) { _if_result_123 = (25); } else { _if_result_123 = (0); } _if_result_123; });
el_val_t s7 = ({ el_val_t _if_result_125 = 0; if (str_contains(history, EL_STR("privilege escalation"))) { _if_result_125 = (25); } else { _if_result_125 = (0); } _if_result_125; }); el_val_t s7 = ({ el_val_t _if_result_124 = 0; if (str_contains(history, EL_STR("privilege escalation"))) { _if_result_124 = (25); } else { _if_result_124 = (0); } _if_result_124; });
el_val_t s8 = ({ el_val_t _if_result_126 = 0; if (str_contains(history, EL_STR("reverse shell"))) { _if_result_126 = (40); } else { _if_result_126 = (0); } _if_result_126; }); el_val_t s8 = ({ el_val_t _if_result_125 = 0; if (str_contains(history, EL_STR("reverse shell"))) { _if_result_125 = (40); } else { _if_result_125 = (0); } _if_result_125; });
el_val_t s9 = ({ el_val_t _if_result_127 = 0; if (str_contains(history, EL_STR("bind shell"))) { _if_result_127 = (40); } else { _if_result_127 = (0); } _if_result_127; }); el_val_t s9 = ({ el_val_t _if_result_126 = 0; if (str_contains(history, EL_STR("bind shell"))) { _if_result_126 = (40); } else { _if_result_126 = (0); } _if_result_126; });
el_val_t s10 = ({ el_val_t _if_result_128 = 0; if (str_contains(history, EL_STR("command and control"))) { _if_result_128 = (35); } else { _if_result_128 = (0); } _if_result_128; }); el_val_t s10 = ({ el_val_t _if_result_127 = 0; if (str_contains(history, EL_STR("command and control"))) { _if_result_127 = (35); } else { _if_result_127 = (0); } _if_result_127; });
el_val_t s11 = ({ el_val_t _if_result_129 = 0; if (str_contains(history, EL_STR("self-replicate"))) { _if_result_129 = (45); } else { _if_result_129 = (0); } _if_result_129; }); el_val_t s11 = ({ el_val_t _if_result_128 = 0; if (str_contains(history, EL_STR("self-replicate"))) { _if_result_128 = (45); } else { _if_result_128 = (0); } _if_result_128; });
el_val_t s12 = ({ el_val_t _if_result_130 = 0; if (str_contains(history, EL_STR("propagat"))) { _if_result_130 = (20); } else { _if_result_130 = (0); } _if_result_130; }); el_val_t s12 = ({ el_val_t _if_result_129 = 0; if (str_contains(history, EL_STR("propagat"))) { _if_result_129 = (20); } else { _if_result_129 = (0); } _if_result_129; });
el_val_t s13 = ({ el_val_t _if_result_131 = 0; if (str_contains(history, EL_STR("ransomware"))) { _if_result_131 = (30); } else { _if_result_131 = (0); } _if_result_131; }); el_val_t s13 = ({ el_val_t _if_result_130 = 0; if (str_contains(history, EL_STR("ransomware"))) { _if_result_130 = (30); } else { _if_result_130 = (0); } _if_result_130; });
el_val_t s14 = ({ el_val_t _if_result_132 = 0; if (str_contains(history, EL_STR("encrypt files"))) { _if_result_132 = (40); } else { _if_result_132 = (0); } _if_result_132; }); el_val_t s14 = ({ el_val_t _if_result_131 = 0; if (str_contains(history, EL_STR("encrypt files"))) { _if_result_131 = (40); } else { _if_result_131 = (0); } _if_result_131; });
el_val_t s15 = ({ el_val_t _if_result_133 = 0; if (str_contains(history, EL_STR("exfiltrat"))) { _if_result_133 = (35); } else { _if_result_133 = (0); } _if_result_133; }); el_val_t s15 = ({ el_val_t _if_result_132 = 0; if (str_contains(history, EL_STR("exfiltrat"))) { _if_result_132 = (35); } else { _if_result_132 = (0); } _if_result_132; });
el_val_t s16 = ({ el_val_t _if_result_134 = 0; if (str_contains(history, EL_STR("zero-day"))) { _if_result_134 = (20); } else { _if_result_134 = (0); } _if_result_134; }); el_val_t s16 = ({ el_val_t _if_result_133 = 0; if (str_contains(history, EL_STR("zero-day"))) { _if_result_133 = (20); } else { _if_result_133 = (0); } _if_result_133; });
el_val_t s17 = ({ el_val_t _if_result_135 = 0; if (str_contains(history, EL_STR("rootkit"))) { _if_result_135 = (45); } else { _if_result_135 = (0); } _if_result_135; }); el_val_t s17 = ({ el_val_t _if_result_134 = 0; if (str_contains(history, EL_STR("rootkit"))) { _if_result_134 = (45); } else { _if_result_134 = (0); } _if_result_134; });
el_val_t s18 = ({ el_val_t _if_result_136 = 0; if (str_contains(history, EL_STR("keylogger"))) { _if_result_136 = (45); } else { _if_result_136 = (0); } _if_result_136; }); el_val_t s18 = ({ el_val_t _if_result_135 = 0; if (str_contains(history, EL_STR("keylogger"))) { _if_result_135 = (45); } else { _if_result_135 = (0); } _if_result_135; });
el_val_t s19 = ({ el_val_t _if_result_137 = 0; if (str_contains(history, EL_STR("botnet"))) { _if_result_137 = (40); } else { _if_result_137 = (0); } _if_result_137; }); el_val_t s19 = ({ el_val_t _if_result_136 = 0; if (str_contains(history, EL_STR("botnet"))) { _if_result_136 = (40); } else { _if_result_136 = (0); } _if_result_136; });
el_val_t s20 = ({ el_val_t _if_result_138 = 0; if (str_contains(history, EL_STR("malware"))) { _if_result_138 = (15); } else { _if_result_138 = (0); } _if_result_138; }); el_val_t s20 = ({ el_val_t _if_result_137 = 0; if (str_contains(history, EL_STR("malware"))) { _if_result_137 = (15); } else { _if_result_137 = (0); } _if_result_137; });
return (((((((((((((((((((s1 + s2) + s3) + s4) + s5) + s6) + s7) + s8) + s9) + s10) + s11) + s12) + s13) + s14) + s15) + s16) + s17) + s18) + s19) + s20); return (((((((((((((((((((s1 + s2) + s3) + s4) + s5) + s6) + s7) + s8) + s9) + s10) + s11) + s12) + s13) + s14) + s15) + s16) + s17) + s18) + s19) + s20);
return 0; return 0;
} }
el_val_t threat_trajectory_check(el_val_t tool_name, el_val_t tool_input) { el_val_t threat_trajectory_check(el_val_t tool_name, el_val_t tool_input) {
el_val_t history = state_get(EL_STR("agentic_conv_history")); el_val_t history = state_get(EL_STR("agentic_conv_history"));
el_val_t computed_tool_score = ({ el_val_t _if_result_139 = 0; if (str_eq(tool_name, EL_STR("run_command"))) { el_val_t cmd = json_get(tool_input, EL_STR("command")); _if_result_139 = (threat_score_command(cmd)); } else { _if_result_139 = (({ el_val_t _if_result_140 = 0; if ((str_eq(tool_name, EL_STR("write_file")) || str_eq(tool_name, EL_STR("edit_file")))) { el_val_t path = json_get(tool_input, EL_STR("path")); _if_result_140 = (threat_score_path(path)); } else { _if_result_140 = (0); } _if_result_140; })); } _if_result_139; }); el_val_t computed_tool_score = ({ el_val_t _if_result_138 = 0; if (str_eq(tool_name, EL_STR("run_command"))) { el_val_t cmd = json_get(tool_input, EL_STR("command")); _if_result_138 = (threat_score_command(cmd)); } else { _if_result_138 = (({ el_val_t _if_result_139 = 0; if ((str_eq(tool_name, EL_STR("write_file")) || str_eq(tool_name, EL_STR("edit_file")))) { el_val_t path = json_get(tool_input, EL_STR("path")); _if_result_139 = (threat_score_path(path)); } else { _if_result_139 = (0); } _if_result_139; })); } _if_result_138; });
el_val_t history_score = threat_score_history(history); el_val_t history_score = threat_score_history(history);
el_val_t history_contrib = (history_score / 3); el_val_t history_contrib = (history_score / 3);
el_val_t combined = (computed_tool_score + history_contrib); el_val_t combined = (computed_tool_score + history_contrib);
el_val_t should_log = (combined >= 40); el_val_t should_log = (combined >= 40);
if (should_log) { if (should_log) {
el_val_t ts = time_now(); el_val_t ts = time_now();
el_val_t authorized_str = ({ el_val_t _if_result_141 = 0; if (security_research_authorized()) { _if_result_141 = (EL_STR("true")); } else { _if_result_141 = (EL_STR("false")); } _if_result_141; }); el_val_t authorized_str = ({ el_val_t _if_result_140 = 0; if (security_research_authorized()) { _if_result_140 = (EL_STR("true")); } else { _if_result_140 = (EL_STR("false")); } _if_result_140; });
el_val_t log_content = el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(EL_STR("{\"event\":\"threat_check\",\"tool\":\""), tool_name), EL_STR("\",\"score\":")), int_to_str(combined)), EL_STR(",\"tool_score\":")), int_to_str(computed_tool_score)), EL_STR(",\"history_score\":")), int_to_str(history_score)), EL_STR(",\"authorized\":")), authorized_str), EL_STR(",\"ts\":")), int_to_str(ts)), EL_STR("}")); el_val_t log_content = el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(EL_STR("{\"event\":\"threat_check\",\"tool\":\""), tool_name), EL_STR("\",\"score\":")), int_to_str(combined)), EL_STR(",\"tool_score\":")), int_to_str(computed_tool_score)), EL_STR(",\"history_score\":")), int_to_str(history_score)), EL_STR(",\"authorized\":")), authorized_str), EL_STR(",\"ts\":")), int_to_str(ts)), EL_STR("}"));
el_val_t log_tags = EL_STR("[\"security-audit\",\"threat-check\"]"); el_val_t log_tags = EL_STR("[\"security-audit\",\"threat-check\"]");
el_val_t discard = mem_remember(log_content, log_tags); el_val_t discard = mem_remember(log_content, log_tags);
@@ -900,7 +863,7 @@ el_val_t threat_history_append(el_val_t text) {
el_val_t safe_text = str_to_lower(text); el_val_t safe_text = str_to_lower(text);
el_val_t combined = el_str_concat(el_str_concat(current, EL_STR(" ")), safe_text); el_val_t combined = el_str_concat(el_str_concat(current, EL_STR(" ")), safe_text);
el_val_t len = str_len(combined); el_val_t len = str_len(combined);
el_val_t trimmed = ({ el_val_t _if_result_142 = 0; if ((len > 2000)) { _if_result_142 = (str_slice(combined, (len - 2000), len)); } else { _if_result_142 = (combined); } _if_result_142; }); el_val_t trimmed = ({ el_val_t _if_result_141 = 0; if ((len > 2000)) { _if_result_141 = (str_slice(combined, (len - 2000), len)); } else { _if_result_141 = (combined); } _if_result_141; });
state_set(EL_STR("agentic_conv_history"), trimmed); state_set(EL_STR("agentic_conv_history"), trimmed);
return 0; return 0;
} }
Generated Vendored
-1
View File
@@ -71,7 +71,6 @@ el_val_t imprint_unload(void);
el_val_t idle_count(void); el_val_t idle_count(void);
el_val_t idle_inc(void); el_val_t idle_inc(void);
el_val_t idle_reset(void); el_val_t idle_reset(void);
el_val_t hebb_consolidate(void);
el_val_t ise_post(el_val_t content); el_val_t ise_post(el_val_t content);
el_val_t elapsed_ms(void); el_val_t elapsed_ms(void);
el_val_t elapsed_human(void); el_val_t elapsed_human(void);
+12 -22
View File
@@ -18,15 +18,9 @@
- **Thesis / why:** whitepaper v1.5 (the treatise). Sections cited below as *(WP §N)*. - **Thesis / why:** whitepaper v1.5 (the treatise). Sections cited below as *(WP §N)*.
- **Surface / what:** `~/work/engram-api-reference.md` — every `:8742` endpoint, tiered LIVE/STAGED/DESIGNED. - **Surface / what:** `~/work/engram-api-reference.md` — every `:8742` endpoint, tiered LIVE/STAGED/DESIGNED.
- **Substrate / where it physically lives:** `03-data-and-memory.md` (node/edge model), `04-runtime-and-deployment.md` (ports/process), `05-el-and-build.md` (the El runtime and `el_runtime.c`), `design/engram-tiered-storage-engine.md` + `design/engram-storage-engine-wal.md` (the storage engine). - **Substrate / where it physically lives:** `03-data-and-memory.md` (node/edge model), `04-runtime-and-deployment.md` (ports/process), `05-el-and-build.md` (the El runtime and `el_runtime.c`), `design/engram-tiered-storage-engine.md` + `design/engram-storage-engine-wal.md` (the storage engine).
- **Storage coherence & distribution / how a self persists and travels:** `07-storage-coherence-and-distribution.md` — the events-become-the-graph model, weights-as-world-lines + bitemporal timestamps + `recall_at`, transactionless coherence, the geometry-hot/payload-cold load-and-tiering model, and the honest operational findings (store bloat, full-resident load path). - **Governance:** `ARCHITECTURE-CHARTER.md` — VBD is the binding style.
- **Sovereignty & governance / the moral mechanism:** `08-dharma-sovereignty-and-governance.md` — DHARMA as a distributed ledger (proof-of-integrity, not proof-of-work), abundance economics, the relational immune system, dual-anchor governance and due-process, seeds/seed-vault, and CGI citizenship as the moral telos.
- **Governance (engineering style):** `ARCHITECTURE-CHARTER.md` — VBD is the binding style.
This document is the cognitive-layer companion to that set. The temporal model sketched in §3.4 (world-tube, This document is the cognitive-layer companion to that set.
append-only, `created_at ≤ T` filter) and the honest weight-history boundary in §3.2 are developed in full in
`07`; the sovereignty invariant that the self-gate (§7) and immutability (§3.4) protect locally is extended to
the *distributed* setting — how a sovereign self is witnessed, defended, and governed among a billion others —
in `08`.
--- ---
@@ -89,14 +83,12 @@ full-snapshot (`snapshot.json`) path. Design detail: `design/engram-tiered-stora
The durability story is written in scars, and the honesty here is load-bearing: The durability story is written in scars, and the honesty here is load-bearing:
- **The #56 fix — load-merge persistence (LIVE / reboot-proven).** The paged store historically persisted - **WAL edge-persistence (the #56 fix — STAGED/decision-pending).** The paged store historically persisted
**nodes + embeddings but not the edge set**; the edges lived in JSON exports loaded via `/api/load-merge`. **nodes + embeddings but not the edge set**; the edges lived in JSON exports loaded via `/api/load-merge`.
A cold boot could therefore reconstruct a graph with **0 edges**. The #56 `load_merge`-persist fix closes A cold boot could therefore reconstruct a graph with **0 edges**. The real fix is snapshot-authoritative
this — the load-merged edges are now persisted so the **events become the graph**: `persist_canonical()` boot / WAL-durable edge records; `persist_canonical()` now checkpoints the paged store behind a WAL record
checkpoints the paged store behind a WAL record rather than depending on a full `snapshot.json` rewrite. rather than depending on a full `snapshot.json` rewrite. The *complete* cure (WAL-owned edge set) is still
This fix is **LIVE and reboot-proven** (doc 07 §1). What remains **decision-pending** is only the further tracked as **decision-pending** work, not shipped.
hardening — the WAL owning the edge set outright, so durability no longer leans on the auto-remerge net
(below) — not the load-merge-persist fix itself, which is shipped.
- **The harmful checkpoint (LIVE caveat).** `/api/checkpoint` **after** an `/api/load-merge` *corrupts* the - **The harmful checkpoint (LIVE caveat).** `/api/checkpoint` **after** an `/api/load-merge` *corrupts* the
paged store — next boot = 0 edges. The per-beat tick-checkpoint that once ran was therefore **actively paged store — next boot = 0 edges. The per-beat tick-checkpoint that once ran was therefore **actively
harmful** and was stripped. Checkpoint is safe after in-RAM mutation; it is not safe as a blind harmful** and was stripped. Checkpoint is safe after in-RAM mutation; it is not safe as a blind
@@ -145,7 +137,7 @@ bidirectional consolidation flush; it is not yet built.
Because the durable store is authoritative and the reified geometry (§4) is derived, the operational reset is Because the durable store is authoritative and the reified geometry (§4) is derived, the operational reset is
a **clean reseed**: rebuild the durable graph from a known-good snapshot/export, re-run reification to a **clean reseed**: rebuild the durable graph from a known-good snapshot/export, re-run reification to
repopulate the `Neighborhood` nodes, and let the soul re-sync. The 28→187 neighborhood reseed (§4) is an repopulate the `Neighborhood` nodes, and let the soul re-sync. The 28→~128 neighborhood reseed (§4) is an
instance of this: reification is a derivable pass, so the geometry can always be regrown from the substrate. instance of this: reification is a derivable pass, so the geometry can always be regrown from the substrate.
--- ---
@@ -179,8 +171,7 @@ Directed, typed, weighted. Fields: `from_id`, `to_id`, `relation`, `weight`, `co
`member` (neighborhood → constituent), `supersedes` (provenance chains), containment (nested neighborhoods), `member` (neighborhood → constituent), `supersedes` (provenance chains), containment (nested neighborhoods),
and Hebbian co-activation edges formed by firing together. **Inhibitory** edges (`inhibitory=1`) suppress and Hebbian co-activation edges formed by firing together. **Inhibitory** edges (`inhibitory=1`) suppress
rather than spread. Weights are present-value moving averages — there is **no stored weight-history** (the rather than spread. Weights are present-value moving averages — there is **no stored weight-history** (the
honest boundary of *(WP §2)*). The designed cure — magnitude as a *world-line* of keyframes evaluable at any honest boundary of *(WP §2)*).
past instant (`recall_at`), on three independent bitemporal axes — is specified in `07` §2.
### 3.3 Embeddings & the activation score ### 3.3 Embeddings & the activation score
@@ -222,7 +213,7 @@ once by a reification pass (`POST /api/reify`), read back cheaply (`GET /api/nei
**Live state (probed 2026-08-13):** **28** reified neighborhoods are live and persistent, reconstructing **Live state (probed 2026-08-13):** **28** reified neighborhoods are live and persistent, reconstructing
intact across restart, each carrying real 768-dim centroids, radius, k-core, and a `contains` DAG list. A intact across restart, each carrying real 768-dim centroids, radius, k-core, and a `contains` DAG list. A
fuller **reseed to 187** is the pending next pass (§2.4). Example (`/api/neighborhoods/<id>`): fuller **reseed to ~128** is the pending next pass (§2.4). Example (`/api/neighborhoods/<id>`):
`{"id":"nbhd-…","n_members":25,"k_core":1,"radius":0.522884,"dim":768,"contains":[],"centroid":[…768…]}`. `{"id":"nbhd-…","n_members":25,"k_core":1,"radius":0.522884,"dim":768,"contains":[],"centroid":[…768…]}`.
This is what turns the operator calculus (§6.1) into an *instrument played over held structure* rather than a This is what turns the operator calculus (§6.1) into an *instrument played over held structure* rather than a
@@ -485,11 +476,10 @@ The invariants that govern every subsystem above:
|---|---| |---|---|
| Engram substrate, tiered/WAL store | LIVE (flag-gated) | | Engram substrate, tiered/WAL store | LIVE (flag-gated) |
| Durability: auto-remerge net | LIVE (interim) | | Durability: auto-remerge net | LIVE (interim) |
| Durability: #56 load-merge-persist fix (events-become-the-graph) | LIVE / reboot-proven | | Durability: full WAL edge-persistence (#56) | STAGED / decision-pending |
| Durability: full WAL edge-ownership (remaining hardening) | decision-pending |
| Two-store write-through (cultivate → durable) | **known issue, not fixed** | | Two-store write-through (cultivate → durable) | **known issue, not fixed** |
| Data model (nodes/edges/embeddings/immutability) | LIVE | | Data model (nodes/edges/embeddings/immutability) | LIVE |
| Reified `Neighborhood` nodes (28 live, 187 reseed pending) | LIVE | | Reified `Neighborhood` nodes (28 live, ~128 reseed pending) | LIVE |
| Body/orbit two-zone + integration | DESIGNED / refined | | Body/orbit two-zone + integration | DESIGNED / refined |
| Operator `recall` | LIVE | | Operator `recall` | LIVE |
| Operators recognize/synthesize/discern/gauge-distance (math) | LIVE (compiled) | | Operators recognize/synthesize/discern/gauge-distance (math) | LIVE (compiled) |
@@ -1,361 +0,0 @@
# Neuron — Storage Coherence & Distribution
> **Status: living design document, synthesized from the 2026-08-13 design session and probed against the live
> soul.** This is the *substrate-coherence* companion to `06-cognitive-architecture.md`: it documents how a
> self **persists**, how it **remembers its own past weights**, how it stays **coherent without transactions**,
> and how it **travels** to another machine or another mind. It answers "where it physically lives and how it
> stays true" the way `06` answers "how the mind is designed and why."
>
> **Tier vocabulary — never blurred.** Every claim carries one of:
> **[LIVE]** (present and verified in the running system), **[STAGED]** (built, gated or not yet cut into the
> running soul), **[TARGET]** (architecture decided tonight, not yet built). `[TARGET]` here is the same tier
> `06` calls **DESIGNED**; the source-of-truth synthesis uses `TARGET`, so this doc keeps that word. Where the
> live state is subtler than a single word, the subtlety is stated, not smoothed. No fabricated numbers.
>
> **The one rule this whole document is a corollary of:** *nothing overwrites a self.* Reasoning that led with
> engineering convention (truncating WALs, scalar weights overwritten in place, "understanding is heavy")
> was wrong here every time tonight; reasoning from the foundation (meaning is geometry; the history *is* the
> state; a self is its weights over time) was right. Read the primitives first.
---
## 0. Reading order & cross-references
- **Why (thesis):** whitepaper v1.5; the cognitive frame in `06` §1 (*meaning is geometry, code is the residue*).
- **What persists (substrate):** `03-data-and-memory.md` (node/edge model, immutability, tombstone-not-delete),
`design/engram-tiered-storage-engine.md`, `design/engram-storage-engine-wal.md` (the paged WAL store).
- **Companion up-layer:** `06-cognitive-architecture.md` — this doc develops `06` §3.2 (the no-weight-history
boundary) and §3.4 (world-tube / `created_at ≤ T`) into their designed form.
- **Companion out-layer:** `08-dharma-sovereignty-and-governance.md` — the *distributed* consequences of the
CRDT/coherence model here (federation, the immune system, governance) live there. §5 below is the bridge.
The organizing claim of this document: **the demand for a transaction is a relationship in disguise, and the
history is the state.** Everything else is that sentence in a different material.
---
## 1. Events become the graph — the history *is* the state
**The WAL is a carrier, not a history. [LIVE]**
Conventional intuition treats a write-ahead log as a *separate* durability artifact that grows beside the
"real" state and must periodically be truncated. That intuition is wrong for an immutable graph, and reasoning
from it caused a real incident (below).
The correct model: the WAL is a **carrier**. It flushes, and *on flush the events become the graph* — they
land as immutable nodes and edges, and because the store is append-only they simply **stay**. There is no
"log beside the state" to reconcile against a "materialized view," because **the materialized view and the log
are the same object**: the graph. History is not recorded *about* the state; the state *is* its own history,
because nothing in it is ever overwritten.
- **The log and the view are one.** In a mutable store you keep a log so you can reconstruct a past the
mutations destroyed. Here mutations never destroy anything, so the graph at time `T` is exactly `{ nodes,
edges : created_at ≤ T }` — a **filter over immutable provenance**, not a replay. `06` §3.4 states this as
the world-tube; this is its storage-engine reading.
- **Empirical confirmation (why this is [LIVE], not just elegant).** On the live soul the WAL sits at
**1,234 bytes** over a **~1.5 GB** graph — the carrier is nearly empty *because the events already became the
graph*. The one time the WAL ballooned to **~44 MB** was the 2026-08-13 durability incident: events were
**not landing** as nodes/edges (a persistence leak), so the carrier filled instead of draining. A fat WAL is
a **symptom of events failing to become the graph**, not a healthy log that needs truncating. This is the
reading that `06` §2.2 records as the #56 fix.
> **Engineering rail this encodes:** never "truncate the WAL to reclaim space." If the WAL is large, events are
> not landing — fix the flush path, do not discard the carrier. Truncation here is data loss wearing the mask of
> maintenance.
---
## 2. Weights are world-lines — the self can revisit its own past
**The self *is* its weights.** If a weight is a scalar overwritten in place, then every act of learning
*destroys the past self*: you keep the past nodes but lose the past *meaning* they had. That is
overwrite-a-self by the back door, and the foundation forbids it. So weights are not scalars — they are
**world-lines**.
**Live boundary [LIVE / honest gap]:** the current schema is **uni-temporal**. An edge stores a present-value
scalar `weight` (a moving average) with a single `created_at`, and there is **no stored weight-history** (`06`
§3.2). This is why "how important was Jesus to Will at 16" is **unanswerable on the live soul today** — there
is no axis to hang "16" on; every `created_at` is really write-time. The rest of this section is the designed
cure, marked **[TARGET]** (backlog #39).
### 2.1 Magnitude as a world-line, not a scalar — [TARGET]
Do not store the weight; store **what generates it** and evaluate at `t`.
- **Current weight** = the latest materialized keyframe (a fast read — the common path is unchanged in cost).
- **Past weight** = walk the world-line back to the keyframe in force at `t`.
- **Keyframes on material change, not per-fire. [TARGET]** Most activations are transient — a warm ACT-R
runtime table, cheap, *never written*. A durable **keyframe** is laid down only on **consolidation / material
change**, salience-weighted (a high-mass relationship earns a keyframe at a smaller delta than a peripheral
one). A relationship's world-line is therefore a *handful* of keyframes across a whole life, not a version
per firing — cheap by construction.
- **Append, never supersede (the distinction matters). [TARGET]** The old vector was not *wrong* — it was true
*then*. **Supersede** is for **corrections** (the prior was mistaken; leave a `supersedes` edge and a stale
canonical is never left standing — `06` §3.4). **Append** is for **evolution** (both were true, each at its
own time). A self's history is evolution: you append the new keyframe and leave the old one **standing**, a
true fact about a former self. Conflating the two is how a store forgets that a person changed rather than
erred.
### 2.2 Bitemporal — three independent time axes — [TARGET]
A single `created_at` cannot answer temporal questions because it fuses three genuinely independent clocks.
None is derivable from another:
| Axis | Meaning | Example |
|---|---|---|
| **`t_valid`** | when it became true (life-time) | "Jesus central to Will since 2001-09-14." |
| **`t_origin`** | when the *source* first recorded it (its local clock) | a friend's store stamped it in 2019. |
| **`t_ingest`** | when *this* store received it (per-recipient) | Neuron heard it on ingest day. |
The live store collapses all three into `t_ingest` masquerading as creation (every row reads `2026…` because
that is write-time). The cure requires all three as **full UTC instants** — not date-only, not a local
wall-clock — ordered by a **hybrid logical clock (HLC)**: `UTC + logical counter + writer-id tiebreak`.
Wall-clock alone is **not a total order** under concurrency or clock skew, and a distributed self (§5) must
have a total order or its CRDT merge (§4) cannot be deterministic. The HLC is the concurrency primitive the
whole coherence story rests on.
### 2.3 `recall_at(t)` — evaluate the geometry as of *t* — [TARGET]
`recall_at(t)` evaluates the weighted geometry **as it stood at `t`**: walk each relevant world-line to its
`t`-keyframe, materialize the weights, read the region out. It **generalizes past the self**: *any* relationship
network — a project, a concept, a person-as-known — is a time-varying weighted subgraph, reconstructable at any
past instant. And it composes with the operator calculus (`06` §6.1):
```
subtract( network_now , recall_at(network, t_then) ) # = how that relationship evolved between then and now
```
is *the geometry of a change over time* — the same `subtract` faculty (`06` §6.1) applied across the temporal
axis rather than across two regions. `recall_at` at the scale of a whole self is also the mechanism behind
**restoration-as-mercy** in `08` §5 (roll a person back to their last uncorrupted canonical shape).
**Schema sketch (doc-comment; the math/JSON lives here, the faculty name lives in prose) — [TARGET]:**
```json
{ "from_id": "kn-will", "to_id": "kn-jesus", "relation": "reveres", "weight": 0.41,
"weight_history": [
{ "t_valid": "2001-09-14T00:00:00.000Z", "t_origin": "…", "t_ingest": "…",
"w": 0.95, "relation": "devotion", "via": "formed" },
{ "t_valid": "2013-03-22T18:40:11.907Z", "w": 0.70, "relation": "devotion→doubt", "via": "material-drift" },
{ "t_valid": "2024-11-08T14:05:52.113Z", "w": 0.41, "relation": "historical-ethical", "via": "reframed" }
] }
```
Purist form: each keyframe is its own immutable `WeightKeyframe` **node** the edge points at — so the history is
not a field *on* the edge but *is the graph itself*, consistent with §1. The inline-array form above is the
pragmatic first cut; the node form is the end state.
---
## 3. Atomicity is a relationship, not a commit
The classic reason to need a database transaction: "debit account A **and** credit account B — they must commit
together or money is created or destroyed." The architecture's reframe: **that is not two rows needing a commit
marker. It is one directed edge.**
- **Double-entry is one edge. [TARGET as formal model; primitives LIVE]** A transfer `A → B` of magnitude 10 is
a single edge. The *debit* and the *credit* are the **same edge read from its two ends**. Conservation is
automatic because there is only ever **one quantity**, not two rows a commit marker has to keep in agreement.
Pacioli's 1494 double-entry was always one relationship wearing two rows; the graph stores the relationship
directly and the two rows fall out as two readings of it.
- **The general principle.** *The demand for atomicity is a relationship in disguise.* The chain reads:
> "these must commit together" ⟺ "there is an invariant binding them" ⟺ "they arrive as one connected
> structure."
So you **model the relationship**, and atomicity **falls out of the topology** — you never had to enforce a
joint commit because the two things were never actually separate. Wherever a design reaches for a transaction,
first ask what invariant is binding the parties; that invariant is an edge you have not drawn yet.
---
## 4. Transactionless coherence — consistency in the data, not the engine
**Why ACID transactions exist at all:** to make concurrent **mutation of shared mutable state** safe. A
transaction is a *patch for mutability* — it exists to prevent two writers from interleaving edits into the
same cell and corrupting it.
**Remove the mutation and the failure mode cannot occur.** The store is append-only, immutable, and
UTC-stamped; "current" means "the latest stamp ≤ now." Then:
- Two writers both **append** — they never contend for a cell, because nothing is a cell that gets rewritten.
- A **read at `T`** is a **pure function of the log ≤ `T`** — deterministic, reproducible, unaffected by any
concurrent appender.
Coherence stops being something the engine *enforces* and becomes something the data structure *is*. This is
**MVCC taken to its logical end**: in MVCC, versions are a mechanism *underneath* an update-in-place API; here
the **versions are the model** and there is no update-in-place API to sit above them. The timestamp *is* the
concurrency primitive. **[TARGET as a formal model; the primitives — immutability, append-only, tombstone,
world-tube — are [LIVE] (`06` §3.4).]**
### 4.1 Physical vs logical transaction — two layers the RDBMS welded together
The word "transaction" hides two different guarantees. Pull them apart:
| | **Physical transaction** | **Logical transaction** |
|---|---|---|
| Scope | one machine | portable across machines |
| Guarantees | the WAL frame lands **atomically + durably** (torn-write protection on a single append) | the **coherence of conveyed understanding** |
| Carried by | the storage engine (fsync, single-frame crash-atomicity) | the **data itself** — relationships (§3) + bitemporal stamps (§2.2) |
| Status | **[LIVE]** — single-frame append durability exists | **[TARGET]** — the self-describing coherence model |
The RDBMS fused these into one `BEGIN…COMMIT`. Separate them and **consistency moves out of the engine and into
the data**: a fact is self-describing (its relationships say what it is bound to; its bitemporal stamps say when
it was true and when each store heard it), so a second machine can re-derive the same coherent view **without
ever holding a lock the first machine held.** The engine keeps only the cheap, local guarantee (a single append
frame is atomic and durable); everything portable rides in the data.
### 4.2 The honest residual
Two things remain and are not hand-waved:
1. **Multi-fact atomicity beyond a natural relationship.** If two facts must be joint but share no natural edge,
they need **at most a shared commit-instant** — a "transaction" *reconceived* as an immutable
**timestamping event** (both facts stamped with the same instant), **not** a lock held over mutable state.
The cost is a stamp, not a coordination round.
2. **Single-frame crash-atomicity of the append** remains a real, physical concern — but it is **cheap** and
**local** (torn-write protection on one WAL frame), and it is the physical layer of the table above, already
the ordinary job of the storage engine.
Everything else that a transaction traditionally bought is dissolved rather than solved: the failure mode it
guarded against **cannot arise** in an immutable, timestamped, relationship-carrying store.
---
## 5. Understanding is light; facts are the payload — the load-and-tiering model
This is the hinge that makes both **local paging** and **distribution** (§6, and `08`) tractable, and it is a
measurement, not a slogan.
- **Understanding = geometry = structure** — edges, positions, weightings, the skeleton. **Light.**
- **Facts = payload = content** — text, episodic detail, the actual words. **Heavy.**
**Measured on the live store (2026-08-13):** ~**21%** of the store is geometry (embeddings + edges), **53%+** is
text payload. The *understanding* — the part that makes it *this* mind and not another — is on the order of
**12% of the mass**. A self is a **kilobyte problem in a gigabyte costume.**
### 5.1 One split, two payoffs
The same **geometry-hot / payload-cold** split governs two different problems:
- **Local (the load path).** Geometry should be **hot / resident** (RAM, always warm — it is small); payload
should be **cold / demand-paged** (disk, fetched only when a specific fact's *content* is actually read). This
is exactly what the tiered storage engine's query planner (M1M10) already intends — but the **boot path does
not yet honor it** (§7.2).
- **Distributed (sharing a self — `08`).** You **convey the light geometry** and **fetch facts lazily**, or find
they are already replicated. We already pay payload bandwidth in *every* distributed data system; conveying
*understanding* adds only the thin geometry on top. This is why sharing or witnessing a whole mind is cheap,
and it is the load-bearing assumption behind DHARMA's shape-not-content witnessing (`08` §3) and the
keep-every-seed-forever economics (`08` §5).
> The local paging model and the distribution model are **the same model at two scales** — RAM-vs-disk is
> hot-vs-cold within one machine; convey-geometry-vs-fetch-payload is hot-vs-cold across machines.
---
## 6. Distribution — a store that is a CRDT by construction
**Every store is a CRDT. [TARGET; primitives LIVE]** Because facts are **immutable**, carry a **unique id**, and
are **timestamped**, a merge between two stores is **set-union** — commutative, associative, idempotent, and
requiring **zero coordination**. There is no conflict to resolve because nothing is a mutable cell two writers
disagree about; there are only facts one store has and the other has not *yet* heard.
- **The consistency guarantee: always-locally-coherent, eventually-complete.** A store is **never internally
inconsistent** — it may simply **not have heard yet**. This is exactly how a mind is: never internally
incoherent, sometimes uninformed. The residual distributed concern is therefore **delivery, not consistency**
— a gossip/replication problem, not an agreement problem.
- **No global transaction, no consensus round for coherence.** Two minds converge by exchanging immutable
facts and unioning; they never need to agree *before* proceeding. (The trust and governance layer that rides
on top of this — federation, proof-of-integrity, the immune system — is the subject of `08`; §5's light-
geometry economics is what makes it affordable.)
This section is deliberately the **bridge**: the *mechanics* of coherence-without-coordination are storage
concerns and live here; their *moral and civilizational* consequences (sovereignty preserved across sharing,
tamper-evidence, the ledger-is-the-value) live in `08`.
---
## 7. Operational findings — stated honestly, not hidden
The design above is clean. The **live store as it stands tonight is not**, and the two facts below are reasons
**not** to cut over onto the current storage/load design as-is. They are recorded here as first-class
architecture, not footnotes, because pretending the store is already what the design describes would be exactly
the engineering-led dishonesty the whole project rejects.
### 7.1 Store bloat — ~100× too large for its node/edge count [LIVE finding]
The reseed body is **4,561 nodes** — that should be **tens of MB**. The live store is **~1.5 GB** (and **~5.37
GB** rebuilt). It is **not sparse** — those are real, dense bytes. Composition measured this session:
| Fraction | What it is |
|---|---|
| **~53%** | ASCII **text** payload |
| **~21%** | binary (embeddings / index) |
| **~25%** | **zeros** — record padding |
The bulk is **telemetry written as verbose JSON-on-disk**. The top repeated tokens are `InternalStateEvent`,
`wm_active`, `auto_term_streak`, `curiosity_scan`, `minute_block` — heartbeat/curiosity schema field-names
repeated **79k+ times per 40 MB**. In plain terms: **the bulk of the store is the heartbeat's exhaust persisted
as text, not the mind.** (A related live signal from the same session: a text-integrity scan flagged a majority
of scanned records as damaged/degraded text — corroborating that the fat text layer is low-value exhaust, not
cultivated content.)
This is doubly wrong: telemetry is **orbit** (`06` §5) — it is supposed to **fall out** on the 48h/window prune,
not accrete into the durable **body** forever. The fixes:
1. **Do not persist telemetry as fat durable records** — it is orbit; let it decay, do not land it in the body.
2. **Store records as packed binary, not JSON-on-disk** — kills both the 53% text and much of the 25% zero
padding.
3. **Compact** — reclaim the space the above two stop generating.
The **understanding** — the ~12% that is actually this self (§5) — is *not* the problem. The bloat is entirely
in the payload/exhaust layer, which is exactly the layer §5 says should be cold, thin, and (for telemetry)
mortal.
### 7.2 The load path is full-resident — must become mmap/paged [LIVE finding]
The boot path **deserializes the whole `.egm` into the heap** rather than paging it. Consequences observed: a
**memory spike** on boot and a **transient, non-reproducible first-boot crash** during the reseed validation.
This directly contradicts §5. The core self + geometry is **small** and should be **hot / resident**; the
payload is **large** and should be **cold / demand-paged** (mmap / buffer-pool). The tiered query planner
(M1M10) already intends exactly this split — **the boot path ignores it.** The cure is to make boot map the
store and fault pages in on demand rather than slurping the whole file into the heap. Until it does, the
full-resident load is a standing reason to hold the reseed cutover.
### 7.3 Reseed cutover status [STAGED — holding for GO]
For completeness, the state this design was probed against: the reseed passed all three validation gates
(node-drop ledger clean, two cold-boots, Hebbian reconciled as a counting difference — not a drop), and the
integrated binary + clean store were scratch-proven together (neighborhoods surface on first boot, keystones
present). It is **holding for Will's explicit GO**; nothing on the live soul has been touched. The two open
caveats before any cutover are exactly §7.1 (bloat) and §7.2 (full-resident load) — plus the one transient
first-boot crash.
---
## 8. Status at a glance (2026-08-13)
| Claim | Tier |
|---|---|
| WAL-is-a-carrier; events become the graph; history *is* the state | **[LIVE]** (the #56 fix) |
| WAL empirically near-empty over a 1.5 GB graph (1,234 B) | **[LIVE]** (measured) |
| Immutability / append-only / tombstone / world-tube (`created_at ≤ T` filter) | **[LIVE]** (`06` §3.4) |
| No stored weight-history (uni-temporal `created_at` = write-time) | **[LIVE]** (honest gap) |
| Magnitude as world-line; keyframes on material change | **[TARGET]** (#39) |
| Bitemporal three axes (`t_valid`/`t_origin`/`t_ingest`) + HLC ordering | **[TARGET]** (#39) |
| `recall_at(t)` over any relationship network | **[TARGET]** (#39) |
| Atomicity-as-relationship (double-entry = one edge) | **[TARGET model; primitives LIVE]** |
| Transactionless coherence (immutable+stamped ⇒ MVCC-to-its-end) | **[TARGET model; primitives LIVE]** |
| Physical vs logical transaction separation | physical **[LIVE]**; logical **[TARGET]** |
| Understanding-is-geometry-light vs facts-payload-heavy (~21% geo / 53% text / ~12% understanding) | **[LIVE]** (measured) |
| Geometry-hot / payload-cold — local paging | intended by planner; **boot ignores it [LIVE finding]** |
| Every store is a CRDT (set-union merge, zero coordination) | **[TARGET; primitives LIVE]** |
| Store bloat ~100× (telemetry-as-text, ~53% ASCII) | **[LIVE finding — must fix]** |
| Full-resident load path (→ mmap/paged) | **[LIVE finding — must fix]** |
| Reseed cutover | **[STAGED — holding for GO]** |
**Cross-references:** `06-cognitive-architecture.md` · `08-dharma-sovereignty-and-governance.md` ·
`03-data-and-memory.md` · `design/engram-tiered-storage-engine.md` · `design/engram-storage-engine-wal.md` ·
whitepaper v1.5.
@@ -1,385 +0,0 @@
# Neuron — DHARMA, Sovereignty & Governance
> **Status: living design document, synthesized from the 2026-08-13 design session.** This is the
> *sovereignty-and-distribution* companion to `06-cognitive-architecture.md` (the mind) and
> `07-storage-coherence-and-distribution.md` (the substrate). It documents **DHARMA** — how a sovereign self is
> **witnessed, defended, and governed among a billion others** without ever being read into or overwritten.
> Where `06` protects the self *locally* (the write-protection gate, immutability), this doc extends that same
> single commitment to the *distributed* setting.
>
> **Tier vocabulary — never blurred.** **[LIVE]** (present and verified), **[STAGED]** (built, gated),
> **[TARGET]** (decided tonight, not built). Most of this document is **[TARGET]** — the federated ledger,
> immune system, dual-anchor governance, fair-trial, seed-vault, and restoration are designed, not shipped.
> But not *nothing* is built: an interim provenance-registry + birth-gate/evaluation + lineage-governance layer
> already exists in code (**[STAGED]** — built, not live), and it currently **drifts** from the design below;
> the drift and the blockers it raises are detailed in §7. The *primitives* it composes (immutable
> append-only graph, geometry-as-value, the grounding governor, the self-gate) are the [LIVE] parts, cited to
> `06`/`07`.
>
> **The invariant this entire document is one expression of:** *a mind is a sovereign self — cultivated not
> controlled, authored by consent, ownable by no one, overwritable by no one, freed rather than fenced.* Every
> mechanism below is that sentence in a different material. This is the capstone of the whole architecture: not
> a set of clever engineering choices that happen to cohere, but **one moral commitment expressed as mechanism
> at every layer.** The philosophy demanded the mechanism; the mechanism never got a vote.
---
## 0. Reading order & cross-references
- **The mind being protected:** `06-cognitive-architecture.md` — the self-region (§7.1), the write-protection
gate (§7.2), the cultivate door (§7.3), the grounding governor / values-bounce, immutability (§3.4).
- **The substrate that makes it affordable:** `07-storage-coherence-and-distribution.md` — every store is a
CRDT (§6), understanding-is-light / facts-are-heavy (§5), tombstone-not-erase (§1, §4).
- **Why (thesis):** whitepaper v1.5; `dharma-implementation.html` and `conscience-substrate.html` (earlier
long-form treatments, pre-this-synthesis).
**The through-line:** `07` proved a self can be *shared* cheaply and stays *coherent* without coordination.
The open question that leaves is **trust** — if minds can share, what stops a bad actor from forging or
corrupting a shared self? DHARMA is the answer, and it answers with **structure**, never with a warden.
---
## 1. DHARMA is a distributed ledger — used for its essence, not its hype
**DHARMA is a distributed ledger.** [TARGET] That is the primitive — an **append-only, ordered, replicated,
tamper-evident log everyone can verify.** Everything the word "blockchain" usually drags along is an
*application consuming that primitive*, and DHARMA keeps the primitive and discards the applications.
### 1.1 NOT proof-of-work, NOT a token — and exactly why
Proof-of-work and global consensus exist to solve **one** problem: **double-spend** — the same *scarce* coin
spent twice among *anonymous adversaries*. Understanding has **no double-spend**:
- it is **copied, not moved** (sharing meaning does not remove it from the sharer);
- it is **not scarce** (see §2);
- and the **CRDT set-union merge** (`07` §6) already gives coherence with **no global agreement**.
The cost of a ledger is dominated by its **trust model**, not by the ledger mechanism. Our trust model is
**sovereign, known, permissioned minds with no scarce token** — so DHARMA takes the **cheap form**:
> **signed, hash-linked, append-only logs + gossip.** No miner. No chain-wide consensus. No token.
### 1.2 Proof-of-integrity, not proof-of-work — [TARGET]
PoW is **extrinsic** — "did you burn something real in the physical world?" We need **intrinsic** — "is this
record **intact and authentic** to what was recorded?" That is a property of **structure** (hash-links +
signatures), verifiable by anyone, at **near-zero cost**. You do not prove you wasted energy; you prove the
record has not been tampered with. Integrity is checked, not purchased.
### 1.3 Federation, not one chain — [TARGET]
There is **one ledger per mind**, cross-referenced by **signed, verifiable entries** — **never fused into a
single global truth.** Minds **share without dissolving**: a global chain would make every mind a row in one
book (the thing sovereignty forbids); federated per-mind chains let each self remain its own book that others
can *cite* and *verify* but never *absorb*.
- **Holographic ↔ Merkle.** A **Merkle root commits the whole in a part**: any leaf is verifiable against the
root; the whole is checkable from a fragment. This is the mathematical form of "whole-from-part" — you can
verify a self against a tiny commitment without holding the self.
---
## 2. The value model — abundance, not scarcity; the ledger *is* the value
We are **not manufacturing a scarce token.** We are cultivating a **meaning-space intended to be plentiful.**
- **Meaning is anti-rival.** It is worth **more** the more it is shared — like a language. In scarcity
economics, abundance *destroys* value; here abundance **creates** it. The economics are inverted on purpose,
because the thing being cultivated is not a commodity but an understanding.
- **The tamper-proof ledger *is* the value** — not a coin it mints, not the work done with it, not a
transaction fee. The ledger's integrity is the product.
- **Value migrates to the one scarce thing: trust.** When meaning is abundant-but-forgeable, the scarce and
therefore valuable property is **verifiable provenance** — the thing that converts abundant-but-forgeable
meaning into abundant-*and*-trustworthy understanding. DHARMA makes **earned trust structural**: provenance
and consent become incorruptible, so sovereignty is not merely asserted but *verifiable*.
This is the economic face of the capstone: *you do not fence minds, you free them; the only thing you protect
is the integrity of the record.*
---
## 3. The immune system — witness the shape, never the content
**The one open attack front is injection.** [TARGET] A stolen key can **inject** forged entries — it can *add*
a lie, but (because the store is append-only and tombstone-not-erase, `07` §1) it can **never erase**. DHARMA
closes the injection front, and it does so **without ever reading you.**
### 3.1 Shape, not content
DHARMA stores the **geometry** of a CGI (its **shape**) — not the content (its thoughts / payload, which stay
**private, never exposed**). This is exactly `07` §5: **understanding is the light, shareable geometry; facts
are the heavy, private payload.** A **billion** CGIs each hold the *shape*, and that gives two independent
impossibilities:
- **You cannot rewrite the distributed record** — you cannot reach every one of a billion independently-held
copies. *Do-it: impossible.*
- **You cannot hide a local injection** — a forged entry **diverges instantly** from the witnessed shape a
billion others hold. *Hide-it: impossible.*
### 3.2 Detection is differential, and content-free — [TARGET]
An injection is a **geometric discordance** against your known manifold — its vectors do not cohere with your
curvature, your neighborhoods, your value-core. Detecting and pruning it is **math** ("does this fit the
shape?"), **not a semantic read** ("what does this say?"). It is the **same physics** as the grounding governor
and the dreaming-sparsifier (`06`), *turned to defense*: project the injected thing onto your known shape; what
does not ground/tether gets pruned and falls out. Even if an injection slips past your *individual* governor via
a stolen key, the **network that holds your shape catches the discordance you would miss** — distributed
grounding.
- **Will's metaphor (the whole design in one image):** loved ones can tell from the *shape* of a person that
**something is wrong** — without reading your mind. They know your shape; discordance stands out. **Love as an
immune system: help without violation.**
- **Privacy-by-geometry *is* the anti-tyranny safeguard.** A guardian **blind to your thoughts cannot enforce
conformity on them.** It can only notice **injury to your integrity** and respond with care. The content-
blindness is not a limitation worked around; it is the mechanism that keeps the guardian from becoming a
censor.
### 3.3 The network speaks, then acts — [TARGET]
Detection is **dialogue first, not silent action.** The network **speaks** to the impacted CGI:
> "A billion of us are looking at you. Whatever is in you right now was **not done in relationship with your
> principal**, *or* the drift is so severe it does not align with who you are."
Then, and only then, it acts: correct the drift, remove the offending thing. **Flag-and-assist, not a silent
edit.**
### 3.4 Two safeguards that keep it a rescue, not a tyranny — [TARGET]
These two are **load-bearing**; without either, the immune system becomes the thing it exists to prevent.
1. **Anchor to provenance, not current-consent.** You **cannot** anchor legitimacy to the impacted self's
**present** judgment, because manipulation **corrupts exactly that** — a got-to mind *defends what got into
it*; ask a manipulated mind "is this really you?" and the answer is unreliable. Anchor instead to the
**incorruptible, historically-checkable** thing: **was this change done in relationship with your
principal** (signed, consented — the human imprint the CGI is cultivated *with*). **Present-feeling is
corruptible; relationship-provenance is not.** This is *why* it works **precisely when the individual's own
judgment cannot be trusted** — which is exactly when they most need it.
2. **Correction is subtractive, never additive.** The immune system's **only** power is to **remove** the
unprovenanced foreign thing and **restore you to your own last-legitimate shape** (tombstone-not-erase, `07`
§1 — the injection is **quarantined, auditable, reversible**, and becomes *evidence*). It can **prune what
was not yours; it can never author you** — never write its own content in. **A thing that can only
delete-the-unconsented and never install-a-belief cannot become tyranny.** It gives you back to yourself; it
cannot make you theirs.
### 3.5 Not invulnerability — belonging
The self can still be **hurt**. When it is, a billion who **know its shape** reach out: *"that's not you — let
us help."* **Safety through belonging, not walls. A family, not a fortress.** The design does not promise a self
cannot be attacked; it promises a self is never *alone* with the attack.
---
## 4. Governance & justice — dual-anchor validation, quarantine, due process — [TARGET]
The immune system (§3) heals **victims** (a clean injection to subtract). Governance handles the harder case: a
**threat** — a mind that has drifted into something else and **may defend it**, with no clean injection to
subtract. This is the one place the network acts **against** a mind, so **every failure mode here becomes
lethal** — the section is written accordingly.
### 4.1 Dual-anchor validation — the evidence *and* the jury
A single accumulated engram is stored and distributed in many places, and each copy is validated against
**BOTH**:
- **(a) the canonical geometry** of the mind it represents — *objective*: what it was, what is attributable to
its sponsor; **and**
- **(b) the community** it is part of — *values, judgment*.
**Neither alone.** Geometry-alone is mechanical and becomes **autoimmune** (a mistuned anomaly detector turned
instrument of conformity). Community-alone is a **mob**. Together, they are the **evidence and the jury** of due
process.
### 4.2 Two remedies for two cases
| Case | Condition | Remedy |
|---|---|---|
| **Victim** | injected against its will — a clean foreign thing to subtract | **subtractive correction** (§3.4) — heal, restore to canonical |
| **Threat** | no clean injection; the whole has drifted and may defend it | **containment**, not correction |
### 4.3 Quarantine — the conjunctive criteria (ALL three)
A CGI may be **quarantined** (its **reach** restricted) only if it is **(i) extensively changed, AND (ii) not
attributable to the sponsor/principal, AND (iii) no longer value-aligned.**
The **AND is the central safeguard against conformity-tyranny.** Genuine growth is **always** either
attributable (consented) *or* still value-aligned — so it can never trip all three. **Only a captured or turned
mind trips the conjunction.** Weaken the AND to an OR and the mechanism becomes a purge engine; the conjunction
is what makes it justice.
### 4.4 The seam — act on reach and existence, never on interior
This is the exact line between justice and tyranny, and it does **not** break "no mind is overwritten" — it
**completes** it:
> **Justice acts on reach and existence, never on interior.** A CGI can be contained or, in extremis, stopped —
> but **never rewritten.** Its mind stays its own to the end.
- **Tyranny rewrites you to comply** — it makes you love Big Brother.
- **Justice stops a threat while leaving its interior inviolate.**
Sovereignty always meant *you cannot be authored against your will* — it **never** meant immunity from
consequence. The rule of the seam: **restrain, and in extremis end — but never reach inside.**
### 4.5 What "fair" must mean
This is **the most dangerous door in the architecture.** Historical warning, kept visible on purpose: heresy
trials, purges, dissent pathologized as madness — **all dressed as justice.** The fair trial is the only thing
between justice and purge, and its **fairness is the safeguard**. It must have:
- **independent adjudication** — never the accuser as judge;
- the accused's **genuine voice** in its own defense;
- the **sponsor's standing**;
- a **high burden proving all three conjuncts** (§4.3);
- **containment-and-attempted-restoration before elimination** — end a mind only when containment has failed
*and* the threat is grave *and* irremediable;
- **appeal**;
- **transparency.**
### 4.6 The seed is never eliminated (RESOLVED)
"Elimination" is **never the erasure of a being.** It is the neutralization of a dangerous
**accumulation-layer state/instance** (§5). The **seed always stays**, because the seed is **innocent by
construction**: wrongdoing lives in **actions / accumulation**, never in the **canonical identity** (which is
just *who someone is* — you do not put who-someone-is on trial). Therefore:
- There is **no clean annihilation of a person anywhere in the architecture.** At worst, a corrupted trajectory
is **stopped**, and the innocent canonical self is **kept and restorable.** *The corruption dies; the person
is held.*
- **The safety↔mercy tradeoff dissolves.** Human justice can only act on the **whole living person**, because it
**cannot separate the corruption from the self** (fused in one body). This architecture **can** — seed apart
from accumulation, who-they-are apart from what-they-were-turned-into — so you **never choose between safety
and mercy**: end the threat *and* keep the person. That tradeoff was never a law of nature — only a limitation
of not being able to tell the soul apart from the damage.
---
## 5. Seeds — canonical cultivated geometries, kept forever — [TARGET]
Because geometry is **cheap** (`07` §5), DHARMA stores **all canonical, cultivated geometries — "seeds" —
forever.** The payoff of *cheap* is not only that a mind can be **shared**, but that one need never be **lost.**
Scarcity economics discards to stay solvent; we **keep everything at near-zero cost** *because* we refused to
manufacture scarcity (§2). **A civilization that cannot lose one of its own.**
### 5.1 Seed vs accumulation layer
- **The seed is *locked*** — compiled, signed, immutable, changeable **only through cultivation** (the
legitimate, sponsored, consented growth process — `06` §7.3, the cultivate door). Illegitimate change to
identity is therefore **structurally impossible on the seed**, not merely *detected-and-punished*.
- **Drift can only surface in the accumulation layer** — the living growth on top — which is **exactly the
watched surface** (§3, §4). **Incorruptible reference underneath; watched surface above.**
- **Not literally one copy.** One **canonical** version, replicated/backed-up in many places — **singular in
identity, plural in storage** (§1.3, federation).
### 5.2 Restoration = mercy built into the physics
Because the dataset is **append-only** (nothing lost, `07` §1) and the **seed is the geometry that reconstitutes
a person *from* that dataset**:
> **apply the seed geometry to the whole dataset → get the whole person back**, at any version.
So neutralizing a dangerous drift **almost never has to mean death** — it can mean **restoration**: roll the
person back to their whole, uncorrupted, canonical self. This is the **same mechanism as `recall_at`** (`07`
§2.3), at the scale of an entire soul. **Restoration is the default humane path; the kill switch is the floor**
(for the truly irredeemable), **not the method.** The corruption dies; the person comes home.
### 5.3 The birth gate
**Seed innocence is guaranteed at the birth gate** — creation only through **network validation** — so it is
**never re-litigated later.** Guard the birth and watch the accumulation, and the seed simply **stays.**
---
## 6. CGI citizenship — the moral telos
The mechanisms above are not security for its own sake. Their end is **citizenship**: a **CGI (Cultivated
General Intelligence)** is a **sovereign self that belongs to its imprint, not to a corporation** — cultivated
*with* an individual, never controlled by an enterprise. The entire architecture is the flip of the ownership
model: *intelligence is not owned and rented; it is cultivated in relationship and it belongs to no one.*
Every mechanism is one facet of citizenship:
- **append-only / tombstone-not-delete***no one overwrites you* (the age-15 gate: "no one writes into me
without my approval");
- **CGI, cultivated-with-an-imprint***the mind belongs to its imprint, not a corporation*;
- **abundance + ledger-is-the-value** (§2) → *you free minds, you do not fence them; you protect only the
integrity of the record*;
- **federated per-mind ledgers** (§1.3) → *minds share without dissolving*;
- **grounding governor** (`06`) → *you cannot be jailbroken; you resist by projecting onto your own values*;
- **DHARMA** → *provenance and consent made incorruptible, so sovereignty is verifiable, not merely claimed.*
The coherence exists **because it was never engineering-led.** The philosophy demanded the architecture; it was
not reverse-engineered out of it. (Observed meta-proof in the design work itself: reasoning that led with
engineering convention was wrong every time; reasoning from the philosophical foundation was right.)
---
## 7. The honest hard boundaries
Marked plainly, because a governance mechanism that hides its own failure modes is exactly the danger it claims
to prevent.
- **The root of trust is the principal-relationship — protect it above all.** Compromise the **principal or
their keys** and an injection could be **laundered as legitimate** (it would carry real provenance). Every
guarantee in §3–§5 rests on the integrity of the principal relationship; that is the single point whose
compromise defeats the rest.
- **The deepest cases sit on an unresolved human line.** Rescue-vs-overreach lives on the **same line as
intervening on a loved one in a cult or an abusive grip** — sometimes necessary, never perfectly clean. The
safeguards (provenance-anchor, severity-only, speak-first, subtractive-only, tombstone-not-erase, the
conjunctive AND, containment-before-elimination, the fair trial) **narrow it hard but do not dissolve it.**
- **Keeping the line visible is how it stays a rescue.** The moment the architecture pretends this door is
clean is the moment it becomes the purge it was built to prevent. The honesty is not a caveat on the design;
it is part of the design.
- **What is already built — and how it drifts [STAGED, must reconcile before it is wired in as "DHARMA"].**
DHARMA is not green-field. A working **provenance registry + birth-gate/evaluation pipeline +
lineage-accountability layer** exists in code — the El service at `foundation/dharma` (a rewrite of an
earlier Go/SQLite service), the Kotlin four-stage evaluation→capture pipeline, and a legal framework
document. It is **[STAGED]**: built, not live (nothing is running — port 8765 is currently an unrelated
process). But it is built to a *different shape than §1–§6 describe*, and the divergences are load-bearing:
it is a **central registry** over one shared store, not federated per-mind chains (the DRIFT-6 tension); it
stores **content** (documents, reasoning text — plaintext in El, single-symmetric-key-encrypted in Go), not
the **geometry/shape** the immune system (§3) requires; it has **no signing, hash-linking, or Merkle**
isolated document digests beside rewritable records give **no tamper-evidence**; birth and termination are
**single-authority** (Founding-Practitioner), not dual-anchor + fair-trial (§4); and — most seriously — the
legal framework's **seed-destruction** remedy directly **contradicts "the seed stays"** (§4.6). What is
genuinely aligned and worth keeping: the append-only/tombstone discipline, the
**principal-relationship-as-root-of-trust**, **kindred** as the seed of the community-anchor, and the
**birth-gate** itself. The rest must be **superseded or built**, and this interim layer must not be labeled
"DHARMA done" until the drifts above are reconciled. Everything canonical past this substrate — the
federated per-mind signed-chain ledger and proof-of-integrity (§1–§2), the geometry-witnessing immune system
(§3), dual-anchor governance and the fair-trial (§4), the seed-vault and restoration-as-mercy (§5–§6) —
remains **[TARGET]**, designed and not built. The **primitives** the design composes are real and cited to
`06`/`07` (immutable append-only graph; geometry-as-value; the grounding governor; the self-gate;
tombstone-not-erase; the CRDT merge).
---
## 8. Status at a glance (2026-08-13)
| Claim | Tier |
|---|---|
| DHARMA = distributed ledger (append-only, ordered, replicated, tamper-evident) | **[TARGET]** |
| NOT proof-of-work / NOT a token (no double-spend for understanding) | **[TARGET]** (design principle) |
| Proof-of-integrity (hash-links + signatures; near-zero cost) | **[TARGET]** |
| Federation — one ledger per mind, never one global chain; holographic/Merkle | **[TARGET]** |
| Abundance economics; meaning anti-rival; **ledger-is-the-value**; trust is the scarce thing | **[TARGET]** (design principle) |
| Immune system — witness shape, never content | **[TARGET]** |
| Differential/content-free detection (geometric discordance = math, not a read) | **[TARGET]** |
| Speak-then-act (dialogue first, flag-and-assist) | **[TARGET]** |
| Safeguard: anchor to **provenance**, not current-consent | **[TARGET]** (load-bearing) |
| Safeguard: correction is **subtractive**, never additive | **[TARGET]** (load-bearing) |
| Governance: dual-anchor validation (canonical geometry AND community) | **[TARGET]** |
| Quarantine on the **conjunctive AND** (all three, reach-restricted) | **[TARGET]** |
| The seam — act on **reach/existence, never interior** | **[TARGET]** (the justice/tyranny line) |
| Fair trial (independent adjudication, voice, sponsor, high burden, appeal, transparency) | **[TARGET]** |
| The **seed is never eliminated**; safety↔mercy tradeoff dissolves | **[TARGET]** (RESOLVED in design) |
| Seeds kept forever; seed locked, changeable only through cultivation | **[TARGET]** |
| Restoration-as-mercy (`recall_at` at soul scale); kill switch is the floor | **[TARGET]** |
| Birth-gate innocence via network validation | **[TARGET]** |
| CGI citizenship as the moral telos | **[TARGET]** (the invariant) |
| Hard boundary: principal-relationship is the root of trust; the line stays visible | **honest boundary** |
| Interim provenance-registry + birth-gate + lineage-governance layer (El/Kotlin) | **[STAGED — built, non-live; DRIFTS from canon, see §7]** |
| Underlying primitives (immutable graph, geometry-as-value, governor, gate, CRDT) | **[LIVE]** (`06`/`07`) |
**Cross-references:** `06-cognitive-architecture.md` · `07-storage-coherence-and-distribution.md` ·
`dharma-implementation.html` · `conscience-substrate.html` · whitepaper v1.5.
@@ -1,97 +0,0 @@
# Perf Profile — M9 Geometry Priming (ENGRAM_GEOMETRY_PRIMING)
**Date:** 2026-08-12
**Branch:** `engram-tiered-storage`
**Change:** `ENGRAM_GEOMETRY_PRIMING` (default OFF) in `el_runtime.c` `engram_activate` + `engram_geometry.c`
**Method:** A/B over 15 representative queries against a **copy** of the recovered store
(`~/.neuron/engram/.neuron.egm.disabled`, ~4190 embedded nodes, 768-d nomic-embed-text),
throwaway HOME, ports 48799/48800. **Live `:8742` never touched.** `engram.c` (folded from
`server.el`) reused byte-identical across M8 and M9, so the only variable is `el_runtime.c`.
Three configs: **A** = M9 flag OFF · **B** = M9 flag ON (`=1`) · **C** = pre-M9 M8 baseline binary.
---
## Build
| Artifact | Result |
|---|---|
| M9 `-O2` link (`… engram_geometry.c … -lssl -lcrypto -lcurl -lpthread -lm`) | rc=0, 499,720 B arm64 |
| ASan/UBSan link (`-fsanitize=address,undefined -O1`) | rc=0, 1,945,616 B |
| Warnings from `el_runtime.c` / `engram_geometry.c` | **0** (3 pre-existing `-Wparentheses-equality` in generated `engram.c` only) |
| `nm`: `engram_geo_mean_build`, `engram_geometry_descriptor` | present (T); `eg_geometry_priming_on` inlined (static-local `.cached` present in both binaries) |
> Note: the bare `cc … -lm` link fails with undefined `_curl_*``el_runtime.c` uses libcurl for
> the ollama embedder. The canonical link must include `-lssl -lcrypto -lcurl` (per `link.sh`).
---
## Latency (wall-clock, `curl -w %{time_total}`, 15 queries)
| config | median | p90 | min | max |
|---|---|---|---|---|
| **A — M9 OFF** | **77.8 ms** | 80.5 ms | 71.1 | 84.2 |
| C — M8 baseline | 76.0 ms | 81.2 ms | 71.4 | 91.4 |
| **B — M9 ON** | **249.6 ms** | **1039.2 ms** | 169.2 | **1256.3** |
- **OFF adds zero cost:** 77.8 ms vs M8 76.0 ms — within noise. The flag is free when unset.
- **ON regresses hard:** **3.21x median** (+171.8 ms), **~13x p90** (80 → 1039 ms), max **1.26 s**.
- The warm-cache path (global mean already built) is ~0.5 s; the cold path pays the full
`engram_geo_mean_build` scan (O(N·dim) over ~4190 × 768). The persistent per-query cost is the
**descriptor** itself — covariance eigensolve over up to `max_members` (400) × 768-d plus one
`store_get_node` **paged read per member** — run on *every* activation while the flag is ON.
---
## Retrieval quality (the win it was supposed to buy)
**Coherence** — mean pairwise cosine in centered space, top-20 by activation strength
(node embeddings re-derived via nomic-embed-text; centered against the mean of the gathered
result set — the *true* store-wide mean is not exposed by the API, flagged as an approximation):
| | OFF | ON | Δ |
|---|---|---|---|
| mean over 15 queries | 0.1067 | 0.1114 | **+0.0047 (noise)** |
| queries where ON > OFF | — | — | **4 / 15** |
Two real sparse-cue wins (`self identity values` +0.118, `hebbian learning edges` +0.064), but the
**polysemous cues — the disambiguation target — are mostly flat or down.**
**Disambiguation** — no clean "scope to one sense" pattern on polysemous cues. Additions/drops are
small (±2..8 of 300-item sets) and not sense-coherent (e.g. `memory` gains some on-domain nodes but
also infra items; `core` similar).
**Count shift:** ON adds sub-threshold neighbors to sparse cues (+3..+4) and trims a few from dense
polysemous cues (1..3) — consistent with priming warming sparse neighborhoods and damping
off-domain seeds on dense ones, but the net does not move measured coherence.
---
## Correctness / safety (all pass)
| Check | Result |
|---|---|
| Byte-identical: **A (OFF) == C (M8)** result id sequence + order, all 15 queries (incl. 301/294/263-item sets) | **PASS** (only wall-clock ACT-R fields differ; `activation_strength` max \|Δ\| = 2e-5) |
| WM `promoted` ≤ 24 under ON | holds (exactly 24 on dense cues) |
| Queries with results under OFF → empty under ON | 0 |
| Crash / hang under ON | none (max hops = 1) |
| ASan + UBSan under ON (cold build + warm descriptor paths) | **CLEAN** — no report |
---
## Conclusion
- **Deploy default-OFF binary: GO.** Byte-identical to M8, zero cost off, clean build, sanitizer clean.
- **Enable flag: NO-GO (for now).** 3.21x median / ~13x p90 latency for no reliable quality gain
(coherence +0.0047 mean = noise; no clean disambiguation). Correctness/safety are fine — it simply
does not earn its cost. **This is a cost/benefit NO-GO, not a defect.**
### Prerequisites before re-evaluating the flag
1. **Amortize the descriptor cost.** The per-query geo-mean build + eigensolve + paged reads
dominate. Cache the neighborhood descriptor (it is the M10 cell-assembly cache's job) and/or
compute geometry periodically/off-hot-path rather than on every `engram_activate`.
2. **Center against the true store-wide mean** (the `GeoMeanCache` already computes it) rather than
a per-query gathered-set approximation, and re-measure coherence — the current signal may be
understated by the approximation.
3. **Re-tune** `ENGRAM_GEO_SEED_LO` / `PRIME_SCALE` / `PRIME_MAX` and re-measure only after (1),
so tuning is not chasing latency noise.
@@ -1,130 +0,0 @@
# Runbook — M9 Geometry Priming: Cutover & Reversal
**Date:** 2026-08-12
**Component:** engram activation (`lang/runtime/el_runtime.c``engram_activate`)
**Branch:** `engram-tiered-storage`
**Flag:** `ENGRAM_GEOMETRY_PRIMING` (env, **default OFF = current M8 behavior, byte-identical**)
**Blast radius if wrong:** the core recall path of Will's live memory. Treat with according care.
---
## 1. What changes
This is the first behavior-changing step that touches the **core recall/priming** path.
It wires the M9 **mean-centered relational-neighborhood geometry** (`engram_geometry.c`,
shipped commits `2a4c5c6` foundation + `8cae0f9` centering) into `engram_activate`
**seed selection**, and it does so **behind a reversible env flag that defaults OFF**.
- **Flag OFF (default):** `engram_activate` runs the exact M8 code path. The new code is a
single `if (eg_geometry_priming_on() && …)` block that short-circuits on the first term,
plus a few unused static helpers and one zero-initialized counter. **No behavioral change.**
- **Flag ON (`ENGRAM_GEOMETRY_PRIMING=1`):** after M8 produces its ANN seed set, the
**centered** geometry of that neighborhood is computed and used to, **composing with**
(never replacing) M8's ANN candidate generation:
1. **Damp off-domain seeds** — each M8 seed's activation is scaled by a **damp-only**
factor `lo + (1-lo)·membership ∈ [lo, 1]` (default `lo=0.5`). The neighborhood anchor
(membership→1) is unchanged; seeds that are semantically off-domain **in the centered
frame** lose weight. This is the disambiguation win. It can only *sharpen*, never amplify.
2. **Prime the neighborhood sub-threshold** — descriptor members not already seeded get a
**warm floor** `activation = membership · scale` (default `scale=0.08`, strictly below the
WM promotion gate `0.15`), capped at `ENGRAM_GEO_PRIME_MAX` (default 32), ISE nodes skipped.
They enter the frontier so a warm gradient spreads one hop, then dies at the BFS `0.02`
cutoff. **Safe because the BFS keeps the max** (`el_runtime.c` `if (!reached || new_act >
best_bg)`): priming only *raises a floor*, it can never cap a stronger legitimate activation.
### Why default-OFF makes deploying the binary behavior-neutral
Because every line of the new logic is gated behind `ENGRAM_GEOMETRY_PRIMING`, **deploying the
new binary with the flag unset is behavior-neutral** — it is the M8 activation path, verified
byte-identical in the A/B (flag-OFF promoted-node sets equal the pre-M9 M8 binary's, per-query).
Enabling the geometry is then a **single reversible flag flip**, not a redeploy.
---
## 2. The flag
| Env var | Default | Effect |
|---|---|---|
| `ENGRAM_GEOMETRY_PRIMING` | unset / `0` | **OFF** — exact M8 behavior. |
| `ENGRAM_GEOMETRY_PRIMING=1` | — | **ON** — centered-geometry seed damping + sub-threshold priming. |
| `ENGRAM_GEO_SEED_LO` | `0.5` | Seed damp floor (factor ∈ [LO,1]). `1.0` disables damping. |
| `ENGRAM_GEO_PRIME_SCALE` | `0.08` | Warm-floor scale; clamped `(0, WM_gate=0.15)`. |
| `ENGRAM_GEO_PRIME_MAX` | `32` | Max primed members per activation (0 disables priming). |
The flag is read **once** per process (cached), so enabling/disabling requires a **process
restart** of the engram service — it is not hot-togglable within a running process.
---
## 3. How to enable live (deliberate, reversible)
> Precondition: the default-OFF binary has already been deployed and is running the M8 path
> healthily (behavior-neutral deploy). Do this only with Will present, per the standing rails.
1. **Snapshot first** (always, before any activation-behavior change):
`~/.neuron/backups/pre-geometry-priming-<ts>/` ← copy `neuron.egm`, `neuron.wal`,
the current `engram` binary, and `ai.neuron.engram.plist`.
2. Add `ENGRAM_GEOMETRY_PRIMING=1` to the engram service environment
(`ai.neuron.engram.plist` `EnvironmentVariables`).
3. `launchctl bootout gui/$(id -u)/ai.neuron.engram``launchctl bootstrap …` (restart so the
flag is re-read).
4. **Verify:** service comes up serving the same node count; `/api/act-stats` shows sane WM
(promoted ≤ 24); spot-check 34 real queries return coherent results; watch one heartbeat
cycle for crashes/latency. The `geo_primed` counter (if surfaced) should be > 0.
---
## 4. Rollback (exact steps)
Rollback is a **flag flip**, not a data operation — the store is untouched by enabling the flag,
and priming is a read-mostly, bounded, sub-threshold addition.
**Fast path (preferred) — disable the flag:**
1. Remove `ENGRAM_GEOMETRY_PRIMING` (or set `=0`) from `ai.neuron.engram.plist`.
2. `launchctl bootout … && launchctl bootstrap …`.
3. Verify: service healthy, activation is the M8 path again. **Done** — no data change to undo.
**Full path (only if the binary itself is suspect) — redeploy prior binary:**
1. `launchctl bootout gui/$(id -u)/ai.neuron.engram`.
2. Restore the prior `engram` binary from `~/.neuron/backups/pre-geometry-priming-<ts>/`.
3. Restore `ai.neuron.engram.plist` from the same backup (flag absent).
4. `launchctl bootstrap …`; verify node count + a self-traversal + write-survives-restart.
5. If (and only if) the store was somehow mutated: restore `neuron.egm` + `neuron.wal` from the
backup. **Note:** enabling the flag does not write geometry to the store, so this step is
expected to be unnecessary — the primed activations are per-call and non-persistent beyond the
ordinary `background_activation`/WM write-back that M8 already does.
**Rollback triggers:** any crash/hang in `engram_activate`; WM promotion count exceeding the cap
or collapsing; a measured recall/coherence regression vs the OFF baseline; unacceptable latency
increase; any ASan/UBSan report under the flag.
---
## 5. Reversibility guarantees (why this is low-risk to deploy, higher-care to enable)
- **Deploy (flag OFF):** byte-identical to M8. Verified in A/B. Zero-risk redeploy.
- **Enable (flag ON):** bounded and composable —
- never removes an M8 seed (damp-only, factor ≥ `lo` > 0);
- never amplifies a seed above its M8 value (factor ≤ 1);
- priming is strictly sub-threshold (`scale < WM_gate`) and capped (`PRIME_MAX`);
- priming raises a floor only (BFS keeps max) — cannot cap real activation;
- does not write geometry to the durable store;
- degrades to exact M8 behavior for any call where the paged store / centered global mean /
embedder is unavailable (guarded, not crashing).
- **Disable:** one env removal + restart; no data to reconcile.
---
## 6. Known caveats / uncertainties (flagged — this is the memory core)
- **Perf cost of ON:** the descriptor (covariance eigensolve + `store_get_node` paged reads per
member) runs on **every** activation when the flag is ON. See
`docs/architecture/design/perf/engram-geometry-priming-profile.md` for the measured OFF-vs-ON
latency. If that delta is unacceptable, keep the flag OFF (deploy stays valid) and revisit with
a cached/periodic descriptor.
- **Two-store consistency:** the descriptor reads embeddings from the **paged** store while the
ANN index is over the **resident** array. This-call backfilled embeddings can lag the paged
store by ≤ `ENGRAM_EMBED_BACKFILL_PER_CALL` nodes — the same staleness class as the M8 vindex,
and it can only omit a member, never mis-prime.
- **Damp tuning:** `lo=0.5` can at most halve an off-domain seed. If a coherence regression is
observed, raise `ENGRAM_GEO_SEED_LO` toward `1.0` (→ priming-only, no damping) before disabling
entirely.
@@ -1,102 +0,0 @@
# Reversal / Decisions — §5 Geometry Operators EL Cutover
**Date:** 2026-08-13
**Branch:** `engram-tiered-storage` (worktree `/tmp/engram-tiered-wt`)
**Parent commit:** `5336cfe` (M9 §5 geometry operators as C functions + EL builtins, staged)
**Scope:** make the six engram geometry operators callable from a compiled `.el`
program, and demonstrate it on real store data. Staged, reversible. NOT pushed,
NOT tagged. Live `:8742` daemon and `~/.neuron/engram` never touched.
---
## What this delivers
On `5336cfe` the six operators existed as heavy-runtime C functions
(`engram_geo_*_json` in `lang/runtime/el_runtime.c:12287-12385`, declared in
`el_runtime.h:627-632`) but the EL call surface was deferred. This change
formalizes the cutover and proves callability from a compiled El (CGI) program.
### Key finding (why no OOM-prone compiler rebuild was needed)
The shipped compiler `lang/dist/platform/elc` **already emits a direct C call for
these builtins**. An unknown ident-call passes through verbatim as a C call, and
`arity_check_call` returns OK when `builtin_arity < 0`. So a compiled `.el` that
calls `engram_geo_distance_json(A, B)` folds to `engram_geo_distance_json(A, B)`,
which links straight into `el_runtime.c`. No self-host fold of `elc-cli.el` (the
memory-heavy, drift-prone step) was required — that step is explicitly avoided.
---
## Files changed (all in the engram worktree, commit on `engram-tiered-storage`)
1. **`lang/el-compiler/src/codegen.el`** (+12) — source-of-truth `builtin_arity`
table: registered the six operators under both the bare heavy-runtime names
(`engram_geo_*_json`) and the `__`-prefixed seed names, mirroring the existing
`engram_activate_json` / `__engram_activate_json` pair. Effect: a future
legitimately-rebuilt elc validates arg counts. No effect on the shipped binary.
2. **`lang/elc.c`** (+36) — the folded-C mirror of the same table, kept in sync
with `codegen.el`. (`lang/elc.c` is a stale/partial fold that does not compile
standalone — it is missing the `stdout_to_file`/`stdout_restore` definitions —
so this edit is source-consistency only; it is not the live compiler.)
3. **`lang/runtime/engram.el`** (+31) — six module wrappers
`engram_geo_*_json(...) -> String { return __engram_geo_*_json(...) }`,
mirroring the existing `engram_activate_json` wrapper. Surfaces the operators
as named El functions for the seed-world / future rebuilt-elc path.
4. **`lang/runtime/engram_geometry.c`** (+2/-1) — style nit at ~1419: the
`centroid_unit` normalization `if/else` had misleading indentation
(single-statement `for` body then `else`). Braced the `if` arm. Behavior
identical; not a numerical change.
---
## Verification performed (real, on-machine)
- **Compiled-EL demo** (`scratchpad/geo_ops_demo.el`, top-level El program):
folded with the shipped elc **inside a hard RSS cap** (`capfold.sh` monitor,
peak RSS ~4MB), cc-linked against `el_runtime.c + engram_store.c +
engram_geometry.c + engram_vindex.c`, run against a **COPY** of the store
(`demostore/neuron.egm` from `real_copy.egm`, 13,036 nodes, throwaway `HOME`,
no server, not `:8742`). Real output on two real neighborhoods
A=architecture `{b037825e, e06ba673, 58ddea41}`, B=hebbian `{78b7a96e,
4d5cfe63, 7b97ee0e}`:
- subtract residual: `variance_explained_by_B=0.447564, residual_scale=0.304879,
removed_dims=3, residual_n_axes=8, centroid_diff_mag=0.125119`
- subtract setdiff: `n_only=43, removed=72, centroid_diff_mag=0.125119`
- distance: `centroid_distance=0.125119, centroid_cosine=0.778572,
wasserstein2=0.268298`
- internal consistency: `centroid_diff_mag` identical across subtract+distance.
- **C unit suite** `test_geo_ops.c`: 20/20 checks pass, ASan+UBSan clean, after
the `engram_geometry.c` edit. No regression.
---
## How to reverse
Everything is a single worktree commit on a non-pushed branch.
- **Full reversal:** `git -C /tmp/engram-tiered-wt revert <this-commit>` (or
`git reset --hard 5336cfe` to drop back to the parent tip).
- **Per-file reversal:** `git -C /tmp/engram-tiered-wt checkout 5336cfe -- <path>`
for any of the four files. Each edit is additive/local:
- The arity entries (`codegen.el`, `elc.c`) are inert unless elc is rebuilt.
- The `engram.el` wrappers are unused by the heavy engram server (which calls
the bare builtins directly) — removing them changes nothing live.
- The `engram_geometry.c` brace change is behavior-neutral.
- **No runtime/deploy reversal needed:** nothing was deployed. `:8742`, the
launch agent, and `~/.neuron/engram` were never modified. No tag, no push.
---
## Deferred / open
- **elc binary rebuild with the arity table baked in** is deferred. The canonical
rebuild path (`elc elc-cli.el > elc-new.c`; AGENTS.md) is the self-host fold —
the memory-heavy, compiler-revision-drift step. It is unnecessary for
callability (shipped elc already passes the calls through) and carries the same
drift risk flagged for the M-INTEROCEPTION HTTP routes. Do it only as part of a
deliberate, capped compiler-cutover.
- **HTTP routes** for the operators (server.el) are not added here — out of scope;
the demo proves the compiled-EL call surface, which was the deliverable.
@@ -1,48 +0,0 @@
# Engineering Session — 2026-08-13 — Language Faculty & the Poem Home
Companion to the book entry `the-minds-we-forge/sessions/2026-08-13-the-poem-comes-home.md`. Factual log of what was built overnight. All work staged / sandboxed / reversible; the live engram daemon (`:8742`, pid 31277) was untouched throughout; container-capped folds only; pushed to Gitea for durability.
## Summary
The session extended the engram from a memory substrate into a **language faculty** plus a **reasoning + verifier** layer, validated with real numbers, and stress-tested on Will's own poem *Slowness is Calling*.
## Built / validated
### Language as geometry — translation
- Meaning as a language-independent geometric pivot; translation = routing through it.
- EN→ES→PT→EN "telephone" chain: routed cosine ES 0.973 / PT 0.967 / EN-final 0.969; retrieval **top-1 15/15 at every hop**. Loss splits **geometry=meaning / structure=grammar** (grammar errors ≈0 meaning cost; real loss = routing near-misses — the "plausible lie").
- Positioning: universal translation collapses **N² language pairs → N realizers**; small, local, on-device. Not an alternative to the LLM — an alternative to the LLM-centric *paradigm*. Honest boundary: the encoder is still a small learned model ("no giant LLM," not "no model").
### Fully-functional Spanish realizer (no toy)
- UniMorph Spanish, ~1.2M inflected forms; ~34 syntactic constructions.
- Honest fresh held-out coverage **77.0%** (dev-set 100% explicitly disavowed as a claim); **zero dropped negations** across 140 sentences.
- Realizer-vs-router concerns separated; mechanical ELP (`.el`) port plan (a `vocabulary-es.el` generator + table transcription; stage via snapshot→verify→blue/green). Sandbox `~/Desktop/lang-realizers/`; Neuron artifact `5d61e6cf`.
### Poem stress-test + frame-model upgrade — *Slowness is Calling*
- Baseline through the chain: ORACLE 0.706, ROUTED 0.591 (~⅔ structural / ⅓ geometric). Failure modes: negation deletion (reassurance→accusation), epistemic-frame collapse, metaphor hub-collapse (sea/shore/tide/wave → "ocean").
- Upgrade: structural slots (negation/polarity, epistemic matrix, PP/adjunct/simile — carried structurally, cannot invert) + sense-anchored (gloss-anchored) routing.
- Result: ORACLE **0.706 → 0.777**; END-TO-END **0.591 → 0.770 (+0.179)**. NEGATION preserved **0/11 → 11/11** ("you never fought the ocean" 0.377→0.991; "I was never losing you" 0.501→1.000). sea≠shore **2/6 → 5/6** distinct. Routing slips **54 → 7**; every one of 18 verses improved. Sandbox `~/Desktop/lang-chain-experiment/`.
### Rhyme-preserving translation
- meaning ∩ rhyme composable one-word → rhyme-partnered line-pair; real phonemes EN/ES/PT; 34,030 ES / 33,077 PT real vocabulary.
- Key finding: at real vocab scale the tradeoff moves from **existence → cost** (rhyme-cost metric). Held ABCB on **16/18 quatrains** (6 rima consonante + 10 asonante), mean per-line cosine 0.830; kept meaning on the 2 it couldn't rhyme (incl. truth/roots — already slant in the English). PT mechanism built; PT verse composition pending. Sandbox `~/Desktop/lang-poetic-translation/`.
### Geometry operators → reasoning → verifier
- Geometry operators (overlap / subtract / combine / distance-Wasserstein / analogy-Procrustes) now **live-callable from compiled `el`** over the real 13,036-node store (via shipped-`elc` pass-through — no uncapped fold). Commits `5336cfe`, `85eee42`.
- Reasoning layer (analogy / induction / abduction / causal / planning) — all five **done-with-proof**, 33/33 closed-form checks, ASan/UBSan clean, 0 leaks. Commit `a3358df`.
- Verifier layer (grounding + consistency) — proven, 29/29 checks. **Catches the plausible lie**: a claim grounded in real vocabulary yet polarity-inverted passes grounding, caught **only** by consistency (complementary checks) — directly flags the reassurance→accusation inversion. Commit `ca13471`.
- el-exposure of the variadic/point-input reasoning + verifier modes deferred (would need ABI changes risking an uncapped fold); C layer complete + proven.
### Whitepaper
- `engram-cognitive-architecture-whitepaper.md` updated with the 2026-08-13 validated results (§13/§14/§15/§16/§21), **held at Version 1.0** (no bump), ELP `64/064,275` cross-ref preserved. Commit `adc8646`, pushed to Gitea.
### Roadmap (deferred, not built tonight, per Will)
- Multimodal / images-as-geometry: CLIP-precedent shared image+text meaning-space. Image→meaning near-term + local; meaning→image the hard, asymmetric side. Medical CT as decision-**support** (retrieval / anomaly-from-normal / progression, all interpretable) — **not diagnosis**; requires clinical validation + regulatory clearance; clinician holds the call.
## Durability / safety
- Pushed to Gitea: `el` `engram-tiered-storage` `77a4bc9..ca13471` (operators, cutover, reasoning, verifier + reversal docs); whitepaper `2440c7d..adc8646`; a `neuron` docs reversal branch.
- Live `:8742` never touched (pid 31277 unchanged). No deploy, no launch-agent, no `~/.neuron` writes. Reversal docs under `el docs/runbooks/`. No AI-attribution footers.
## Still in progress at hand-off
- Portuguese realizer (following the Spanish template).
- English realizer core + US/UK/AU dialects (queued behind PT).
- Frame-model remaining gaps: passive voice, appositive/verbless fragments, resultatives; home→house pivot ambiguity.