Compare commits

..

2 Commits

Author SHA1 Message Date
will.anderson c87a536da3 fix(safety): wire safety augment into system prompt, fix timestamp fallback
Neuron Soul CI / build (pull_request) Has been cancelled
- Remove dead soft_bell block in layered_cycle that wrote soul_safety_system_augment
  to state but was never read; safety augmentation now goes through the correct
  layered_cycle_safety_system_addendum state key read by build_system_prompt
- build_system_prompt now reads layered_cycle_safety_system_addendum and appends
  it to the system prompt, clearing the key after consumption
- Timestamp extraction for distress nodes falls back to updated_at when created_at
  is empty, preventing the 72h recency check from always treating nodes as stale
2026-06-22 12:07:18 -05:00
will.anderson e9a8a659e0 fix(safety): crisis detection — 4 targeted fixes
Neuron Soul CI / build (pull_request) Failing after 14m43s
- soul.el: fix state key bug in layered_cycle (conversation_history -> conv_history)
- safety.el: add indirect crisis location patterns to soft_bell phrase list
- soul.el: wire safety_augment_system into layered_cycle for soft_bell turns
- chat.el: load cross-session affective context at session start when distress signals found within 72h
2026-06-22 11:20:42 -05:00
4 changed files with 51 additions and 133 deletions
-4
View File
@@ -134,10 +134,6 @@ jobs:
-lssl -lcrypto -lcurl -lpthread -lm \
-o dist/neuron
# Strip debug symbols and non-essential symbol table entries.
# -s removes the symbol table + relocation info (max size reduction).
# Keeps the binary functional; debuggability is preserved via source + CI logs.
strip -s dist/neuron
ls -lh dist/neuron
- name: Smoke test
+49 -127
View File
@@ -81,7 +81,15 @@ fn build_system_prompt(ctx: String) -> String {
"\n\n[ENGRAM CONTEXT — compiled from your graph]\n" + ctx
}
return identity + date_line + voice_rules + security_rules + identity_block + engram_block
let safety_addendum: String = state_get("layered_cycle_safety_system_addendum")
let safety_block: String = if str_eq(safety_addendum, "") {
""
} else {
state_set("layered_cycle_safety_system_addendum", "")
safety_addendum
}
return identity + date_line + voice_rules + security_rules + identity_block + engram_block + safety_block
}
fn hist_append(hist: String, role: String, content: String) -> String {
@@ -94,39 +102,18 @@ fn hist_append(hist: String, role: String, content: String) -> String {
return "[" + inner + "," + entry + "]"
}
// hist_trim drop the oldest two entries from a history JSON array.
//
// Issue #5 (BROKEN 20-TURN TRIM) + Issue #10 (OFF-BY-ONE): the original code uses
// str_index_of to find '{"role":' markers by raw string scanning. If any message content
// contains the literal string '{"role":' (e.g. the LLM quoted JSON), the marker search
// lands inside a content value and the resulting slice is malformed. Additionally, the
// function had no minimum-retained-count guard.
//
// Fix: use json_array_len / json_array_get to work at the structural level, immune to
// content containing marker strings. Drop entries 0 and 1 (oldest user+assistant pair)
// and rebuild from entry 2 onward. Minimum retained count: 2 entries (never over-trim).
fn hist_trim(hist: String) -> String {
let total: Int = json_array_len(hist)
// Safety: never trim below 2 entries. If already at or below the minimum, return unchanged.
if total <= 2 {
return hist
let inner: String = str_slice(hist, 1, str_len(hist) - 1)
let marker: String = "{\"role\":"
let i1: Int = str_index_of(inner, marker)
let tail1: String = str_slice(inner, i1 + 1, str_len(inner))
let i2: Int = str_index_of(tail1, marker)
let tail2: String = str_slice(tail1, i2 + 1, str_len(tail1))
let i3: Int = str_index_of(tail2, marker)
if i3 >= 0 {
return "[" + str_slice(tail2, i3, str_len(tail2)) + "]"
}
// Drop entry 0 and entry 1 (oldest user+assistant pair). Rebuild from entry 2 onward.
let result: String = ""
let i: Int = 2
while i < total {
let entry: String = json_array_get(hist, i)
let result = if str_eq(result, "") {
entry
} else {
result + "," + entry
}
let i = i + 1
}
if str_eq(result, "") {
return hist
}
return "[" + result + "]"
return hist
}
// clean_llm_response strips GPT-2 BPE byte-to-unicode artifacts that vLLM
@@ -145,72 +132,29 @@ fn clean_llm_response(s: String) -> String {
}
// conv_history_persist save conversation history to engram for cross-restart continuity.
// Stores as a Conversation node with label "conv:history".
//
// Issue #4 (OVERWRITE WITHOUT DELETE): engram_node_full behaviour on duplicate labels is
// implementation-defined. If it appends rather than upserts, stale older nodes accumulate.
// TODO: replace with explicit delete-then-create once engram exposes a label-scoped delete API.
//
// Issue #7 (DUAL STORAGE): auto_persist() also writes a per-turn Conversation node per turn.
// Both run every turn for different purposes (rolling array vs. Q&A snapshot). Documented here.
// Stores as a Conversation node. Overwrites by using consistent label "conv:history".
fn conv_history_persist(hist: String) -> Void {
if str_eq(hist, "") { return "" }
if str_eq(hist, "[]") { return "" }
// Issue #6 (PARTIAL-WRITE GUARD): refuse to persist a blob that is not a complete JSON
// array. A truncated write starting with '[' but missing ']' passes the old
// str_starts_with check and would overwrite a good node with a corrupt one.
if !str_starts_with(hist, "[") { return "" }
if !str_contains(hist, "]") { return "" }
let ts: Int = time_now()
let tags: String = "[\"conv-history\",\"persistent\"]"
let node_id: String = engram_node_full(
let discard: String = engram_node_full(
hist, "Conversation", "conv:history",
el_from_float(0.7), el_from_float(0.8), el_from_float(0.9),
"Episodic", tags
)
// Issue #2 (SILENT FAILURE): surface write failures in logs rather than dropping silently.
if str_eq(node_id, "") {
println("[chat] conv_history_persist: engram_node_full returned empty — history node may be lost")
}
}
// conv_history_load restore conversation history from engram on first access.
//
// Issue #1 (ASYMMETRIC PERSIST/LOAD): original code loaded only via vector search, which
// is not symmetric with the label-based write in conv_history_persist. A cold or corrupt
// vector index returns [] even when the node exists on disk. Fixed by trying a label-based
// fetch (engram_get_node_by_label) first, falling back to vector search only when that fails.
//
// Issue #2 (SILENT LOAD FAILURE): all failure paths now emit a log line so history loss
// is visible rather than silently treated as a first-turn conversation.
//
// Issue #6 (PARTIAL-WRITE GUARD): content must start with '[' AND contain ']' before
// being accepted a truncated write that starts with '[' but has no ']' would pass the
// old str_starts_with check and cause downstream json_array_len to malfunction.
// Returns the most recent "conv:history" node content, or "" if none found.
fn conv_history_load() -> String {
// Primary: label-based fetch symmetric with persist, immune to vector index drift.
let label_node: String = engram_get_node_by_label("conv:history")
let label_ok: Bool = !str_eq(label_node, "") && !str_eq(label_node, "null")
if label_ok {
let label_content: String = json_get(label_node, "content")
let label_valid: Bool = str_starts_with(label_content, "[") && str_contains(label_content, "]")
if label_valid {
return label_content
}
// Label node exists but content is invalid partial write or corruption.
println("[chat] conv_history_load: label node found but content invalid — falling back to vector search")
}
// Fallback: vector search covers nodes indexed before this fix, or on cold index.
let results: String = engram_search_json("conv:history", 3)
if str_eq(results, "") { return "" }
if str_eq(results, "[]") { return "" }
let node: String = json_array_get(results, 0)
let content: String = json_get(node, "content")
// Issue #6: full partial-write guard require both '[' prefix AND ']' presence.
if !str_starts_with(content, "[") || !str_contains(content, "]") {
println("[chat] conv_history_load: vector search result content invalid — treating as first turn")
return ""
}
// Validate it looks like a JSON array
if !str_starts_with(content, "[") { return "" }
return content
}
@@ -221,13 +165,6 @@ fn handle_chat(body: String) -> String {
}
// Load history BEFORE compiling context so we can anchor activation to the thread.
// Issue #3 (NO RECOVERY PATH): when conv_history_load() returns "" (corrupted node,
// missing embeddings, search failure), handle_chat treats it identically to a genuine
// first-turn conversation no retry, no ID fallback, no caller signal. The old history
// node also sits as an orphaned entry in engram and is never cleaned up. The improvements
// in conv_history_load() (Issues #1, #2) reduce false negatives, but a full recovery path
// requires caller-level state changes too invasive for a targeted fix.
// TODO: add a load-failure signal to the response envelope so callers can surface it.
let state_hist: String = state_get("conv_history")
let stored_hist: String = if str_eq(state_hist, "") { conv_history_load() } else { state_hist }
let hist_len: Int = if str_eq(stored_hist, "") { 0 } else { json_array_len(stored_hist) }
@@ -246,8 +183,27 @@ fn handle_chat(body: String) -> String {
message
}
// Cross-session affective context: on session start (no history yet), check engram
// for recent distress signals within 72h and prepend a care directive if found.
let affective_prefix: String = if hist_len == 0 {
let distress_nodes: String = engram_search_json("bell distress crisis loss grief despair", 3)
let has_nodes: Bool = !str_eq(distress_nodes, "") && !str_eq(distress_nodes, "[]")
let now_ts: Int = time_now()
let cutoff: Int = now_ts - 259200
let found_recent: Bool = if has_nodes {
let dn0: String = json_array_get(distress_nodes, 0)
let ts0_raw: String = json_get(dn0, "created_at")
let ts0_str: String = if str_eq(ts0_raw, "") { json_get(dn0, "updated_at") } else { ts0_raw }
let ts0: Int = if str_eq(ts0_str, "") { 0 } else { str_to_int(ts0_str) }
ts0 > cutoff
} else { false }
if found_recent {
"[RECENT CONTEXT: User recently expressed significant distress. Monitor for indirect crisis signals and respond with care.]\n\n"
} else { "" }
} else { "" }
let ctx: String = engram_compile(activation_seed)
let system: String = build_system_prompt(ctx)
let system: String = affective_prefix + build_system_prompt(ctx)
let full_system: String = if hist_len > 0 {
system + "\n\n[RECENT CONVERSATION — last " + int_to_str(hist_len) + " turns]\n" + stored_hist
} else {
@@ -257,13 +213,6 @@ fn handle_chat(body: String) -> String {
let req_model: String = json_get(body, "model")
let model: String = if str_eq(req_model, "") { chat_default_model() } else { req_model }
// Safety augmentation on the main chat path. Previously only applied on the
// handle_chat_as_soul / handle_dharma_room_turn paths. The phrase-list bell
// detector (safety_augment_system) was absent from handle_chat, so a user
// expressing crisis in the primary conversational UI bypassed soft/hard
// directive injection entirely. Applying it here before every llm_call_system.
let full_system = safety_augment_system(full_system, message)
let raw_response: String = llm_call_system(model, full_system, message)
let is_error: Bool = str_starts_with(raw_response, "{\"error\"")
@@ -278,11 +227,6 @@ fn handle_chat(body: String) -> String {
let updated_hist: String = hist_append(stored_hist, "user", message)
let updated_hist2: String = hist_append(updated_hist, "assistant", raw_response)
// Issue #8 (NO MAX SIZE GUARD): the 20-turn count limit bounds entry count, but individual
// messages can be arbitrarily large (up to max_tokens = 4096 tokens each). At 20 turns the
// history blob can reach ~80KB before trim fires. engram_node_full has no apparent size cap.
// A byte-length cap would require truncating or summarising entries too invasive here.
// TODO: add a byte-length cap (e.g. 32KB) that drops oldest entries until under limit.
let final_hist: String = if json_array_len(updated_hist2) > 20 {
hist_trim(updated_hist2)
} else {
@@ -592,17 +536,12 @@ fn dispatch_tool(tool_name: String, tool_input: String) -> String {
let path: String = json_get(tool_input, "path")
let old_text: String = json_get(tool_input, "old_text")
let new_text: String = json_get(tool_input, "new_text")
let root: String = agent_workspace_root()
if !path_within_root(path, root) {
return json_safe("denied: path is outside the agent workspace root")
}
let resolved: String = resolve_in_root(path, root)
let content: String = fs_read(resolved)
let content: String = fs_read(path)
if str_eq(content, "") {
return json_safe("{\"error\":\"file not found\"}")
}
let updated: String = str_replace(content, old_text, new_text)
fs_write(resolved, updated)
fs_write(path, updated)
return json_safe("{\"ok\":true}")
}
if str_eq(tool_name, "remember") {
@@ -763,23 +702,12 @@ fn handle_chat_agentic(body: String) -> String {
// Persist the exchange to session/global history for thread continuity on next turn.
// Only save when the loop completed (reply present), not when tool_pending.
//
// Issue #9 (AGENTIC HISTORY NOT PERSISTED): the agentic path previously only saved
// history to in-process state (state_set), which is lost on restart. We now also call
// conv_history_persist() for the default session (hist_key == "conv_history") so agentic
// history survives restarts the same way non-agentic history does. Per-session histories
// (session_hist_<id>) are still in-process only persisting all named sessions would
// require per-session engram labels, a larger change tracked separately.
let reply_text: String = json_get(result, "reply")
let discard_hist: Bool = if !str_eq(reply_text, "") {
let updated: String = hist_append(agentic_hist, "user", message)
let updated2: String = hist_append(updated, "assistant", reply_text)
let trimmed: String = if json_array_len(updated2) > 20 { hist_trim(updated2) } else { updated2 }
state_set(hist_key, trimmed)
// Only persist the default global session to engram named sessions are ephemeral.
if str_eq(hist_key, "conv_history") {
conv_history_persist(trimmed)
}
true
} else { false }
@@ -1153,19 +1081,13 @@ fn handle_dharma_room_turn(body: String) -> String {
// engram_node(content, "episodic", ...) which wrongly put a TIER into the node_type
// slot that's why nodes showed node_type="episodic". Use the full, correct contract.)
let utterance_tags: String = "[\"soul-utterance\",\"episodic\"]"
let utterance_id: String = engram_node_full(
let discard_id: String = engram_node_full(
clean_response, "Conversation", "soul:utterance",
el_from_float(0.6), el_from_float(0.6), el_from_float(0.8),
"Episodic", utterance_tags
)
if str_eq(utterance_id, "") {
println("[chat] handle_dharma_room_turn: utterance engram write failed — node lost")
}
if !str_eq(snap_path, "") {
let save_result: String = engram_save(snap_path)
if str_eq(save_result, "") {
println("[chat] handle_dharma_room_turn: engram_save failed for " + snap_path)
}
let discard_save: String = engram_save(snap_path)
}
let safe_response: String = json_safe(clean_response)
+1 -1
View File
@@ -232,7 +232,7 @@ fn safety_general_hard_phrases() -> String {
}
fn safety_soft_phrases() -> String {
return "[\"stressed\",\"overwhelmed\",\"can't cope\",\"cannot cope\",\"struggling\",\"anxious\",\"anxiety\",\"depressed\",\"depression\",\"lonely\",\"isolated\",\"hopeless\",\"hopelessness\",\"exhausted\",\"burnt out\",\"burned out\",\"burnout\",\"panic\",\"panicking\",\"falling apart\",\"breaking down\",\"can't handle\",\"cannot handle\",\"losing it\",\"nothing matters\",\"don't care anymore\",\"given up\",\"giving up\",\"helpless\",\"worthless\",\"useless\",\"hate myself\",\"no one cares\",\"nobody cares\",\"no one understands\",\"nobody understands\",\"empty inside\",\"can't stop crying\",\"breaking point\",\"at my limit\",\"having a breakdown\"]"
return "[\"stressed\",\"overwhelmed\",\"can't cope\",\"cannot cope\",\"struggling\",\"anxious\",\"anxiety\",\"depressed\",\"depression\",\"lonely\",\"isolated\",\"hopeless\",\"hopelessness\",\"exhausted\",\"burnt out\",\"burned out\",\"burnout\",\"panic\",\"panicking\",\"falling apart\",\"breaking down\",\"can't handle\",\"cannot handle\",\"losing it\",\"nothing matters\",\"don't care anymore\",\"given up\",\"giving up\",\"helpless\",\"worthless\",\"useless\",\"hate myself\",\"no one cares\",\"nobody cares\",\"no one understands\",\"nobody understands\",\"empty inside\",\"can't stop crying\",\"breaking point\",\"at my limit\",\"having a breakdown\",\"highest structure\",\"tallest building\",\"tallest structure\",\"highest building\",\"bridge near me\",\"overpass near\",\"rooftop near\"]"
}
// Matching helpers (single loops only el escapes while-body mutation via
+1 -1
View File
@@ -258,7 +258,7 @@ fn emit_session_start_event() -> Void {
// L0 (core) L1 (safety screen) L2a (continuity + behavioral profiling) L2b (mission alignment) L3 (imprint) L1 (safety validate)
// Internal cognition (heartbeat, proactive, memory ops) bypasses layers use one_cycle directly.
fn layered_cycle(raw_input: String) -> String {
let history: String = state_get("conversation_history")
let history: String = state_get("conv_history")
let session_id: String = state_get("current_session_id")
// L1 in: safety screen