feat(recall): recall-completeness #48

Merged

will.anderson merged 1 commits from improve/recall-recall-completeness into main 2026-06-22 18:24:23 +00:00

Conversation 0 Commits 1 Files Changed 1

+329 -92

will.anderson commented 2026-06-22 17:54:54 +00:00

Owner

Copy Link

Copy Source

Memory recall improvements for recall-completeness

Memory recall improvements for recall-completeness

will.anderson added 15 commits 2026-06-22 17:55:00 +00:00

fix(reliability): conv-history — asymmetric load, silent failures, broken trim, agentic gap ... 615f0cee08

Issues addressed:
- #1 ASYMMETRIC PERSIST/LOAD: conv_history_load() now tries engram_get_node_by_label()
  first (symmetric with the label-based write), falling back to vector search only when
  label lookup returns nothing. Immune to cold/corrupt vector index.
- #2 SILENT LOAD FAILURE: all failure paths in conv_history_load() and conv_history_persist()
  now emit a println log line rather than silently returning "" or dropping writes.
- #3 NO RECOVERY PATH: documented as TODO with explanation of why a full recovery path
  (retry, ID fallback, orphan cleanup) is too invasive for a targeted fix here.
- #4 OVERWRITE WITHOUT DELETE: documented with TODO to replace engram_node_full with
  explicit delete-then-create once engram exposes a label-scoped delete API.
- #5/#10 BROKEN TRIM / OFF-BY-ONE: hist_trim() rewritten to use json_array_len /
  json_array_get (structural JSON ops) instead of raw str_index_of scanning for
  '{"role":' markers. Immune to marker strings appearing inside message content.
  Minimum retained count guard added: never trims below 2 entries.
- #6 PARTIAL-WRITE GUARD: conv_history_persist() refuses to write a blob that doesn't
  contain both '[' and ']'. conv_history_load() requires both before accepting content.
- #7 DUAL STORAGE: documented with a comment at the persist call site.
- #8 NO MAX SIZE GUARD: documented as TODO with rationale for why a byte-length cap
  requires a more invasive change (entry truncation or summarisation).
- #9 AGENTIC HISTORY NOT PERSISTED: handle_chat_agentic() now calls conv_history_persist()
  for the default global session (hist_key == "conv_history") after updating state,
  matching the non-agentic path's durability. Named sessions remain in-process only.

fix(reliability): cross-session-affective ... 34551695a1

- Fix state key mismatch: soul.el layered_cycle now reads conv_history
  (not conversation_history), unblocking the safety_score_distress_history
  history-amplification path in safety_threat_score
- Add safety_augment_system call on the main handle_chat path so the
  phrase-list bell detector fires on all chat turns, not just dharma rooms
- Add cross-session affective engram query in load_identity_context() at
  boot; stores distress/crisis signals from prior sessions under
  soul_affective_context with a 7-day soft recency filter

fix(reliability): engram-write — guard all fire-and-forget writes ... 494d973a3b

Every engram_node_full call that dropped its return value now binds it
and emits a println on empty string. engram_save calls in consolidate,
heartbeat, and dharma-room-turn are checked for failure. The two API
handlers (log_state_event, tune_config) that skipped api_persisted()
now match the read-back-after-write contract used everywhere else in
neuron-api.el.

Files changed:
- chat.el: conv_history_persist, handle_dharma_room_turn, auto_persist
- soul.el: emit_session_start_event, seed_persona_from_env HTTP check
- memory.el: mem_save, mem_boot_count_inc
- neuron-api.el: handle_api_log_state_event, handle_api_tune_config,
  handle_api_consolidate (engram_save + session summary write)
- awareness.el: ise_post local-engram fallback path

TODO comments added for non-atomic patterns (issues #12, #13) and
the missing circuit breaker (#14) — these require new primitives.

fix(reliability): safety-resilience — bell augmentation, safe mode, dedup logging, tab escaping, handle_chat coverage deddb9a18e

fix(reliability): safety-resilience — bell augmentation, safe mode, dedup logging, tab escaping, handle_chat coverage aa70c5dde6

fix(reliability): engram-connection ... d008649c3e

- entrypoint.sh: extend engram health-check timeout 30->60s; set
  EL_HTTP_TIMEOUT_MS=10000 and EL_HTTP_CONNECT_TIMEOUT_MS=3000 to bound
  awareness loop blocking window to 10s/call (down from 60s default)
- soul.el: 3-attempt retry loop for boot-time /api/nodes+/api/edges fetch;
  validate non-empty JSON array before loading to prevent silent zero-node
  identity graph from transient post-healthcheck network hiccup
- awareness.el: soft circuit-breaker in ise_post (opens after 3 failures,
  30s backoff, half-open probe); /api/sync refresh skips HTTP call when
  breaker is open; error-JSON detection on sync response

TODOs: full async dispatch, connection pooling (require EL futures/persistent curl)

fix(reliability): llm-retry — empty response detection, configurable max_tokens, connector timeout ... 47d0e6f985

Issue #5: detect empty string from llm_extract_text() as an error in handle_chat,
handle_chat_as_soul, and handle_dharma_room_turn. The C runtime silently returns ""
when the LLM response content array is missing or all blocks fail to parse; without
this guard the empty string passes through to callers as a silent empty reply.

Issue #9: make agentic_loop max_tokens configurable via NEURON_LLM_MAX_TOKENS env
var (default 4096). The hardcoded value is marginal for long tool chains (8 iterations
x 4096 tokens); operators can now set 8192+ for complex multi-step tasks without
rebuilding. Non-agentic path (llm_call_system) still uses the C runtime hardcode —
that fix lives in el_runtime.c (see TODO block added in this commit).

Issue #10: increase connector_tools_json and tool_auto_approved curl --max-time from
2s to 5s to reduce false-empty tool lists when neuron-connectd is under transient
load. Graceful degradation to [] on bridge down is unchanged.

Issues #1/#2/#3/#4/#6/#8: documented as TODO comments in chat.el. These require
targeted C runtime changes in el_runtime.c (llm_provider_request retry loop,
EL_LLM_TIMEOUT_MS separation, HTTP 429 backoff, 5xx retry, EL_HTTP_MAX_RESPONSE_BYTES
cap). Architectural decisions recorded so they are traceable to root causes.

fix(reliability): route-error-recovery ... 2865d6ad26

- Issue #3: err_404/err_405 now emit HTTP 404/405 via __status__ envelope instead of HTTP 200
- Issue #4: add auth_check() function to handle_request; enforces NEURON_TOKEN on all routes except /health and /lineage
- Issue #5: missing required params now return HTTP 400 (__status__ envelope) in /api/chat (GET+POST), /imprint/contextual, /imprint/user, and handle_chat
- Issue #6: LLM unavailable in handle_chat now returns HTTP 503 instead of HTTP 200
- Issue #7: add 32 KB message size guard on POST /api/chat before engram_compile and LLM
- Issue #8: add TODO comment to route_health documenting the live-engram-query problem and the /health/deep split plan
- Issue #9: add comment to hist_trim documenting fragile str_index_of parser and silent data corruption risk
- Issue #10: add TODO comment in handle_request documenting missing per-IP rate limiting
- Issue #11: fix connectd_post temp file collision — add monotonic sequence counter so concurrent requests get unique paths
- Issue #12: fix call_mcp_bridge fixed temp file race — add monotonic sequence counter for unique paths under concurrent load
- Issues #1/#2: add TODO comment in handle_request documenting EL no-exception limitation and SIGSEGV handler gap

fix(reliability): replace undefined session_exists with session_get check 392d2416ec

fix(reliability): conversation history 94b71b6e6b

fix(reliability): cross-session affective state f35569d4bb

fix(reliability): engram write db33b0cb91

fix(reliability): engram connection 575ff1329a

fix(reliability): route error recovery e447a87a00

fix(reliability): LLM retry 6edf9937dd

will.anderson added 1 commit 2026-06-22 18:11:59 +00:00

feat(recall): recall-completeness improvements ... 18e040acb1

- Lower engram_compile_ranked threshold 25->15: include moderately-relevant older nodes
- Extend sentinel cleanup from _sel_9 to _sel_14 to prevent JSON noise
- Add engram_split_topics for multi-topic decomposition (AND/and/also/plus)
- Add engram_extract_entities for named entity dedicated searches
- Add engram_detect_recall_intent for boosted 40-candidate search on recall phrases
- Add engram_is_continuation replacing brittle 50-char threshold (now 80 + pronoun/opener detection)
- Add engram_compile_multi with depth 8 (was 5) and 30-candidate search pool
- Add engram_nodes_merge for clean two-array deduplication
- Replace engram_compile with multi-topic/entity/recall-boost version; budget 6000->8000
- Safe JSON truncation: scan for last } before budget cap instead of raw str_slice
- handle_chat and agentic_chat: use engram_is_continuation; thread snip 150->250
- session_preload: add project-status and session-summary search queries

will.anderson merged commit b6052f9de3 into main 2026-06-22 18:24:23 +00:00

Sign in to join this conversation.

Reviewers

No Reviewers

Labels

No items

No labels

Milestone

No items

No Milestone

Projects

Clear projects

No projects

Assignees

Clear assignees

tim.lingo (Tim Lingo) will.anderson (Will Anderson)

No Assignees

1 Participants

Notifications

Subscribe

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: neuron-technologies/neuron#48