chore(dist): update soul.c with PR #63/#65/#66 + Task 1 chat.el changes

Manually adds compiled C equivalents for:
- distill_transcript() — last-3-messages extractor; wires into
  handle_dharma_room_turn and handle_dharma_room_turn_agentic
- current_engine_note() — appended to system prompt in handle_chat
  so Neuron can answer 'what model am I running on?' truthfully (PR #66)
- llm_base_url / llm_wire_format / json_escape / openai_chat_complete —
  OpenAI-compatible provider path in handle_chat_agentic (PR #65)
- flag_true() — tolerant agentic flag check (PR #63)

Compile verified: 6 pre-existing warnings, 0 errors.

awareness.el

@@ -219,9 +219,14 @@ fn proactive_curiosity() -> Bool {
     // Activate each term independently so substring seed-finding hits many nodes.
     // hops=1 (not 2): the in-process Engram has grown to 165K+ nodes. hops=2 BFS
     // visits far more nodes and returns much larger JSON blobs. On a graph this
-    // large, hops=1 still activates all directly-related nodes AND triggers the
-    // semantic seed supplement (cosine sim ≥ 0.70 scan over all embedded nodes),
-    // giving broad working-memory coverage without the quadratic blowup of hops=2.
+    // large, hops=1 still activates all directly-related nodes, giving broad
+    // working-memory coverage without the quadratic blowup of hops=2.
+    //
+    // NOTE: a semantic seed supplement (cosine sim ≥ 0.70 scan over embedded nodes)
+    // was planned alongside hops=1 but is NOT yet implemented — embed_ok in
+    // heartbeats confirms Ollama is reachable, but no embedding call is made during
+    // activation. The seed-finding loop in el_runtime.c uses istr_contains only.
+    // (2026-06-30 self-review: corrected stale comment)
     let curiosity_seed: String = curiosity_term_a + " " + curiosity_term_b + " " + curiosity_term_c
     let results_a: String = engram_activate_json(curiosity_term_a, 1)
     let results_b: String = engram_activate_json(curiosity_term_b, 1)
@@ -278,11 +283,20 @@ fn proactive_curiosity() -> Bool {
     let safe_auto: String = str_replace(auto_term, "\"", "'")
 
     let wmc: Int = engram_wm_count()
+    // wm_top snapshot in curiosity_scan ISE: top-3 WM nodes by weight.
+    // Heartbeat already records top-5 every 60s; curiosity_scan fires every 30s
+    // (scan_ms = beat_ms/2) and is the PRIMARY activation driver during idle.
+    // Without wm_top here, we can't see which nodes actually entered WM after
+    // each curiosity round — only the aggregate count. Top-3 is enough to
+    // diagnose "stuck on X" patterns without bloating the ISE payload.
+    // (2026-07-01 self-review)
+    let wm3: String = engram_wm_top_json(3)
     let ise: String = "{\"event\":\"curiosity_scan\",\"seed\":\"" + curiosity_seed
         + "\",\"auto_term\":\"" + safe_auto
         + "\",\"minute_block\":" + int_to_str(minute_block)
         + ",\"activated\":" + int_to_str(total_found)
         + ",\"wm_active\":" + int_to_str(wmc)
+        + ",\"wm_top\":" + wm3
         + ",\"ts\":" + int_to_str(ts) + "}"
     ise_post(ise)
     return total_found > 0

chat.el

@@ -594,6 +594,44 @@ fn engram_compile(intent: String) -> String {
     if str_starts_with(ctx, "[") { return truncated + "]" }
     return truncated
 }
+// distill_transcript — extract the salient tail from a full conversation transcript.
+//
+// Purpose: before activating working memory on a transcript, reduce it to the
+// last N turns. Activating on the ENTIRE transcript (which may contain hundreds
+// of messages) would produce noisy, over-broad seed finding — too many nodes match
+// too many words, collapse the WM to breakthrough-floor nodes. Taking only the tail
+// focuses activation on what's contextually live right now.
+//
+// Handles two transcript formats:
+//   JSON array: [{"role":"human","content":"..."},...]  → extract last 3 messages' content
+//   Plain text: raw string                              → return last 500 chars
+//
+// Returns a string of at most 500 chars suitable for engram_compile/engram_activate.
+// (Added 2026-07-01 self-review: was called in handle_dharma_room_turn and
+// handle_dharma_chat but never defined — caused build failure since June 30.)
+fn distill_transcript(transcript: String) -> String {
+    if str_eq(transcript, "") { return "" }
+    // JSON array format: extract last 3 messages' content fields
+    if str_starts_with(transcript, "[") {
+        let n: Int = json_array_len(transcript)
+        if n == 0 { return "" }
+        let m0: String = json_array_get(transcript, n - 1)
+        let m1: String = if n > 1 { json_array_get(transcript, n - 2) } else { "" }
+        let m2: String = if n > 2 { json_array_get(transcript, n - 3) } else { "" }
+        let c0: String = json_get(m0, "content")
+        let c1: String = json_get(m1, "content")
+        let c2: String = json_get(m2, "content")
+        let combined: String = c2 + " " + c1 + " " + c0
+        let len: Int = str_len(combined)
+        if len > 500 { return str_slice(combined, len - 500, len) }
+        return combined
+    }
+    // Plain text: return last 500 chars
+    let len: Int = str_len(transcript)
+    if len > 500 { return str_slice(transcript, len - 500, len) }
+    return transcript
+}
+
 fn json_safe(s: String) -> String {
     let s1: String = str_replace(s, "\\", "\\\\")
     let s2: String = str_replace(s1, "\"", "\\\"")
@@ -1238,6 +1276,86 @@ fn agentic_api_key() -> String {
     return env("NEURON_LLM_0_KEY")
 }
 
+// ── OpenAI-compatible providers (Ollama / OpenAI / Grok / Gemini) ──────────────────────────────
+// The brain speaks Anthropic's Messages format by default. When the active provider uses the
+// OpenAI-compatible wire format (NEURON_LLM_0_FORMAT=openai) with a configured base URL
+// (NEURON_LLM_0_URL, e.g. http://localhost:11434/v1 for local Ollama), basic chat turns are served
+// here instead of the Anthropic agentic loop.
+// v1 SCOPE: plain chat completion only — NO tools / agentic loop yet (that is a follow-up port).
+// This block is ADDITIVE: the Anthropic path is untouched and stays the default.
+
+fn llm_base_url() -> String {
+    return env("NEURON_LLM_0_URL")
+}
+
+fn llm_wire_format() -> String {
+    let f: String = env("NEURON_LLM_0_FORMAT")
+    if str_eq(f, "") {
+        return "anthropic"
+    }
+    return f
+}
+
+// Escape a decoded string so it can be embedded back into a JSON string literal.
+fn json_escape(s: String) -> String {
+    let a: String = str_replace(s, "\\", "\\\\")
+    let b: String = str_replace(a, "\"", "\\\"")
+    let c: String = str_replace(b, "\n", "\\n")
+    let d: String = str_replace(c, "\r", "\\r")
+    return d
+}
+
+// Basic (non-agentic) chat completion against an OpenAI-compatible endpoint.
+// [safe_sys] is already JSON-escaped; [messages_json] is the same JSON array the Anthropic path
+// builds (e.g. [{"role":"user","content":"..."}]). Returns the soul's standard {"reply":"..."}.
+fn openai_chat_complete(model: String, base_url: String, api_key: String, safe_sys: String, messages_json: String) -> String {
+    // Prepend the system prompt as an OpenAI "system" message, then the existing turn array.
+    let inner: String = if json_array_len(messages_json) > 0 {
+        str_slice(messages_json, 1, str_len(messages_json) - 1)
+    } else {
+        ""
+    }
+    let msgs: String = if str_eq(inner, "") {
+        "[{\"role\":\"system\",\"content\":\"" + safe_sys + "\"}]"
+    } else {
+        "[{\"role\":\"system\",\"content\":\"" + safe_sys + "\"}," + inner + "]"
+    }
+    let req_body: String = "{\"model\":\"" + model + "\""
+        + ",\"max_tokens\":4096"
+        + ",\"messages\":" + msgs
+        + "}"
+
+    let h: Map = {}
+    map_set(h, "content-type", "application/json")
+    // Ollama needs no key; OpenAI / Grok / Gemini use a Bearer token.
+    if !str_eq(api_key, "") {
+        map_set(h, "Authorization", "Bearer " + api_key)
+    }
+
+    let url: String = base_url + "/chat/completions"
+    let raw_resp: String = http_post_with_headers(url, req_body, h)
+
+    let is_error: Bool = str_starts_with(raw_resp, "{\"error\"") || str_contains(raw_resp, "\"error\":")
+    if is_error {
+        return "{\"error\":\"llm unavailable\",\"reply\":\"\"}"
+    }
+
+    // Parse OpenAI response shape: choices[0].message.content
+    let choices: String = json_get_raw(raw_resp, "choices")
+    let eff_choices: String = if str_eq(choices, "") {
+        "[]"
+    } else {
+        choices
+    }
+    if json_array_len(eff_choices) < 1 {
+        return "{\"error\":\"empty response\",\"reply\":\"\"}"
+    }
+    let first: String = json_array_get(eff_choices, 0)
+    let message: String = json_get_raw(first, "message")
+    let content: String = json_get(message, "content")
+    return "{\"reply\":\"" + json_escape(content) + "\",\"tools_used\":[]}"
+}
+
 fn agentic_tools_literal() -> String {
     return "[" +
         "{\"name\":\"read_file\",\"description\":\"Read contents of a file from disk.\",\"input_schema\":{\"type\":\"object\",\"properties\":{\"path\":{\"type\":\"string\",\"description\":\"Absolute file path\"}},\"required\":[\"path\"]}}," +
@@ -1802,7 +1920,14 @@ fn handle_chat_agentic(body: String) -> String {
 
     // Use caller-supplied session_id if provided, otherwise generate a bridge id.
     let session_id: String = if str_eq(req_session, "") { next_bridge_id() } else { req_session }
-    let result: String = agentic_loop(session_id, model, safe_sys, tools_json, messages, h, "")
+    // Provider fork: OpenAI-compatible providers (Ollama/OpenAI/Grok/Gemini) take the plain-completion
+    // path (v1, no tools); everything else stays on the Anthropic agentic loop (the default).
+    let use_openai: Bool = !str_eq(llm_base_url(), "") && str_eq(llm_wire_format(), "openai")
+    let result: String = if use_openai {
+        openai_chat_complete(model, llm_base_url(), agentic_api_key(), safe_sys, messages)
+    } else {
+        agentic_loop(session_id, model, safe_sys, tools_json, messages, h, "")
+    }
 
     // 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.

council/README.md

@@ -0,0 +1,123 @@
+# Neuron Council Service
+
+Anti-confabulation layer for the Neuron soul. Before a claim enters long-term memory, the council convenes: three independent LLMs vote on whether the claim is plausible, uncertain, or a confabulation. The aggregate vote produces a confidence score and tags that downstream storage can act on.
+
+## Running the service
+
+```bash
+# Foreground
+python3 council_service.py --port 7771
+
+# Background (managed by LaunchAgent on macOS)
+launchctl load ~/Library/LaunchAgents/ai.neuron.council.plist
+launchctl unload ~/Library/LaunchAgents/ai.neuron.council.plist
+```
+
+Logs: `~/.neuron/logs/council.log`
+
+## API
+
+### `POST /api/neuron/council/verify`
+
+```json
+// Request
+{ "claim": "...", "context": "..." }
+
+// Response
+{
+  "id": "550e8400-e29b-41d4-a716-446655440000",
+  "claim": "...",
+  "confidence": 0.85,
+  "council_votes": ["plausible", "plausible", "plausible"],
+  "summary": "3/3 council members agree this is plausible.",
+  "tags": ["verified"],
+  "latency_ms": 1420
+}
+```
+
+### `GET /healthz`
+
+Returns `{"status": "ok"}` when the service is up.
+
+## Confidence thresholds and tag meanings
+
+| Votes plausible | Confidence | Tags |
+|---|---|---|
+| 3/3 | 0.85 | `verified` |
+| 2/3 | 0.65 | `council-split` |
+| 1/3 or 0/3 | 0.30 | `unverified`, `council-flagged` |
+| Ollama down | 0.50 | `council-unavailable` |
+
+Recommended storage policy:
+- `confidence >= 0.65` → store normally
+- `0.30 <= confidence < 0.65` → store with `council-split` tag for later review
+- `council-flagged` → store in a quarantine bucket or reject entirely
+- `council-unavailable` → store normally (fail-open); council will re-evaluate later
+
+## How to call from soul (.el)
+
+The soul is implemented in Neuron's Emacs Lisp-like `.el` language. Add a pre-storage hook in the memory capture path:
+
+```elisp
+;; In memory.el or safety.el — pre-storage council check
+(defun council-verify (claim context)
+  "Call the council service. Returns a plist with :confidence and :tags."
+  (let* ((url "http://localhost:7771/api/neuron/council/verify")
+         (body (json-encode `((claim . ,claim) (context . ,context))))
+         (resp (neuron-http-post url body))
+         (data (json-decode resp)))
+    data))
+
+;; In the capture handler — wire it in before (engram-write ...)
+(defun capture-memory-with-council (claim context &rest store-args)
+  (let* ((verdict (council-verify claim context))
+         (confidence (plist-get verdict :confidence))
+         (tags (plist-get verdict :tags)))
+    (when (>= confidence 0.30)  ; only reject hard confabulations if you want
+      (apply #'engram-write
+             (append store-args
+                     (list :council-confidence confidence
+                           :council-tags tags))))))
+```
+
+The exact hook point depends on where `engram-write` (or equivalent) is called in `memory.el`. Search for the write call and wrap it with `capture-memory-with-council`.
+
+## Future soul.c patch point
+
+If the soul is ever rewritten in C or another compiled language, the integration point is:
+
+```c
+// Before inserting a memory node into the engram database:
+CouncilResult result = council_verify(claim, context);
+if (result.confidence < COUNCIL_REJECT_THRESHOLD) {
+    log_warn("Council flagged claim as confabulation (conf=%.2f): %s",
+             result.confidence, claim);
+    return MEMORY_REJECTED;
+}
+memory_node.council_confidence = result.confidence;
+memory_node.council_tags = result.tags;
+engram_insert(memory_node);
+```
+
+## Council members
+
+The council is currently three models:
+- `neuron:latest` — the primary Neuron model
+- `dolphin3:8b` — uncensored general-purpose model for independent perspective
+- `neuron-ft:latest` — fine-tuned Neuron variant
+
+Each member votes independently with a 10-second timeout. If a member times out, their vote counts as "uncertain". If Ollama is entirely unreachable, the service returns `council-unavailable` immediately (fail-open: confidence 0.5, no rejection).
+
+## Example curl
+
+```bash
+# Should get high confidence (true fact)
+curl -s http://localhost:7771/api/neuron/council/verify -X POST \
+  -H 'Content-Type: application/json' \
+  -d '{"claim": "Neuron is a personal AI memory system built by Will Anderson", "context": "product description"}'
+
+# Should get low confidence (false claim)
+curl -s http://localhost:7771/api/neuron/council/verify -X POST \
+  -H 'Content-Type: application/json' \
+  -d '{"claim": "The Eiffel Tower is located in Berlin and was built in 1950", "context": "geography"}'
+```

council/compressor_service.py

@@ -0,0 +1,234 @@
+#!/usr/bin/env python3
+"""
+Neuron CCR Phase 1 — System Prompt Compressor Service.
+
+Receives a verbose soul system prompt and returns a semantically equivalent
+but token-dense compressed version. Reduces system prompt tokens by 60-80%
+with no behavioral information loss.
+
+Architecture reference: foundation/forge/docs/token-compression-architecture.md
+Model: qwen3:1.7b (primary), neuron:latest (fallback)
+
+Usage:
+    python3 compressor_service.py [--port 7772]
+
+API:
+    POST /api/neuron/compress
+    {"system_prompt": "...", "context_type": "identity|rules|memory"}
+
+    Response:
+    {"compressed": "...", "original_tokens": N, "compressed_tokens": N,
+     "reduction_pct": X, "model": "...", "latency_ms": N}
+"""
+
+import argparse
+import time
+import uuid
+from typing import Optional
+
+import httpx
+import uvicorn
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+from pydantic import BaseModel
+
+# ---------------------------------------------------------------------------
+# Config
+# ---------------------------------------------------------------------------
+
+OLLAMA_BASE = "http://localhost:11434/api/generate"
+
+# qwen3:1.7b is the architecture-specified compressor (Phase 1).
+# neuron:latest is the fallback: already running, domain-appropriate.
+PRIMARY_MODEL = "qwen3:1.7b"
+FALLBACK_MODEL = "neuron:latest"
+MODEL_TIMEOUT = 60.0  # seconds; compression of a long prompt can take time
+
+# Compression prompt — preserves all facts/rules/constraints, strips verbosity.
+# /no_think suppresses qwen3's chain-of-thought tokens, keeping output clean.
+COMPRESSOR_PROMPT_TEMPLATE = """\
+/no_think
+You are a semantic compression engine. Compress the following system prompt while preserving ALL specific facts, rules, constraints, and named entities. Do not lose any information that would change behavior. Output ONLY the compressed text, nothing else.
+
+Original prompt:
+{system_prompt}
+
+Compressed (preserve all facts and rules):"""
+
+# ---------------------------------------------------------------------------
+# App
+# ---------------------------------------------------------------------------
+
+app = FastAPI(
+    title="Neuron Compressor Service",
+    description="CCR Phase 1 — system prompt compression for the Neuron soul",
+    version="1.0.0",
+)
+
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+
+# ---------------------------------------------------------------------------
+# Models
+# ---------------------------------------------------------------------------
+
+class CompressRequest(BaseModel):
+    system_prompt: str
+    context_type: Optional[str] = "mixed"  # identity | rules | memory | mixed
+
+
+class CompressResponse(BaseModel):
+    id: str
+    compressed: str
+    original_tokens: int
+    compressed_tokens: int
+    reduction_pct: float
+    model: str
+    context_type: str
+    latency_ms: int
+
+
+# ---------------------------------------------------------------------------
+# Token estimation (rough: word_count × 1.3, matching architecture doc)
+# ---------------------------------------------------------------------------
+
+def estimate_tokens(text: str) -> int:
+    """Rough token count estimate: words × 1.3. No tokenizer dependency."""
+    words = len(text.split())
+    return max(1, int(words * 1.3))
+
+
+# ---------------------------------------------------------------------------
+# Core compression
+# ---------------------------------------------------------------------------
+
+async def ollama_available(client: httpx.AsyncClient) -> bool:
+    """Quick connectivity check to Ollama."""
+    try:
+        await client.get("http://localhost:11434/", timeout=2.0)
+        return True
+    except (httpx.ConnectError, httpx.TimeoutException):
+        return False
+
+
+async def compress_with_model(
+    client: httpx.AsyncClient, model: str, prompt_text: str
+) -> str:
+    """
+    Call a single Ollama model to compress the given text.
+    Returns the compressed string, or "" on failure.
+    """
+    payload = {
+        "model": model,
+        "prompt": prompt_text,
+        "stream": False,
+        # Keep temperature low for deterministic compression
+        "options": {
+            "temperature": 0.1,
+            "top_p": 0.9,
+        },
+    }
+    try:
+        resp = await client.post(OLLAMA_BASE, json=payload, timeout=MODEL_TIMEOUT)
+        resp.raise_for_status()
+        data = resp.json()
+        return data.get("response", "").strip()
+    except (httpx.TimeoutException, httpx.HTTPStatusError, Exception):
+        return ""
+
+
+async def run_compression(system_prompt: str, context_type: str) -> CompressResponse:
+    start = time.monotonic()
+    request_id = str(uuid.uuid4())
+
+    original_tokens = estimate_tokens(system_prompt)
+    prompt_text = COMPRESSOR_PROMPT_TEMPLATE.format(system_prompt=system_prompt)
+
+    async with httpx.AsyncClient() as client:
+        # Connectivity gate
+        if not await ollama_available(client):
+            latency_ms = int((time.monotonic() - start) * 1000)
+            return CompressResponse(
+                id=request_id,
+                compressed=system_prompt,  # passthrough on failure
+                original_tokens=original_tokens,
+                compressed_tokens=original_tokens,
+                reduction_pct=0.0,
+                model="unavailable",
+                context_type=context_type,
+                latency_ms=latency_ms,
+            )
+
+        # Try primary model (qwen3:1.7b), fall back to neuron:latest
+        compressed = await compress_with_model(client, PRIMARY_MODEL, prompt_text)
+        model_used = PRIMARY_MODEL
+
+        if not compressed:
+            compressed = await compress_with_model(client, FALLBACK_MODEL, prompt_text)
+            model_used = FALLBACK_MODEL
+
+        if not compressed:
+            # Both models failed — passthrough
+            latency_ms = int((time.monotonic() - start) * 1000)
+            return CompressResponse(
+                id=request_id,
+                compressed=system_prompt,
+                original_tokens=original_tokens,
+                compressed_tokens=original_tokens,
+                reduction_pct=0.0,
+                model="both-failed",
+                context_type=context_type,
+                latency_ms=latency_ms,
+            )
+
+    compressed_tokens = estimate_tokens(compressed)
+    reduction_pct = round(
+        (1.0 - compressed_tokens / max(1, original_tokens)) * 100.0, 1
+    )
+    latency_ms = int((time.monotonic() - start) * 1000)
+
+    return CompressResponse(
+        id=request_id,
+        compressed=compressed,
+        original_tokens=original_tokens,
+        compressed_tokens=compressed_tokens,
+        reduction_pct=reduction_pct,
+        model=model_used,
+        context_type=context_type,
+        latency_ms=latency_ms,
+    )
+
+
+# ---------------------------------------------------------------------------
+# Routes
+# ---------------------------------------------------------------------------
+
+@app.post("/api/neuron/compress", response_model=CompressResponse)
+async def compress(req: CompressRequest):
+    return await run_compression(req.system_prompt, req.context_type or "mixed")
+
+
+@app.get("/healthz")
+async def health():
+    return {"status": "ok", "service": "compressor", "version": "1.0.0"}
+
+
+# ---------------------------------------------------------------------------
+# Entrypoint
+# ---------------------------------------------------------------------------
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Neuron Compressor Service (CCR Phase 1)")
+    parser.add_argument("--port", type=int, default=7772, help="Port to listen on")
+    parser.add_argument("--host", default="127.0.0.1", help="Host to bind to")
+    args = parser.parse_args()
+
+    print(f"[compressor] Starting on {args.host}:{args.port}")
+    print(f"[compressor] Primary model: {PRIMARY_MODEL}")
+    print(f"[compressor] Fallback model: {FALLBACK_MODEL}")
+    uvicorn.run(app, host=args.host, port=args.port, log_level="info")

council/council_service.py

@@ -0,0 +1,224 @@
+#!/usr/bin/env python3
+"""
+Neuron Council Service — LLM anti-confabulation layer.
+
+Fires 3 parallel Ollama calls and aggregates votes to produce a
+confidence score + tags for any claim before it enters memory.
+
+Usage:
+    python3 council_service.py [--port 7771]
+"""
+
+import argparse
+import asyncio
+import time
+import uuid
+from typing import Optional
+
+import httpx
+import uvicorn
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+from pydantic import BaseModel
+
+# ---------------------------------------------------------------------------
+# Config
+# ---------------------------------------------------------------------------
+
+OLLAMA_BASE = "http://localhost:11434/api/generate"
+COUNCIL_MODELS = ["neuron:latest", "dolphin3:8b", "neuron-ft:latest"]
+MODEL_TIMEOUT = 45.0  # seconds per model (models may need to load from cold)
+
+SYSTEM_PROMPT_TEMPLATE = """\
+You are a fact-checker. You will be given a claim.
+Your job: assess if it is accurate, internally consistent, and grounded in reality.
+Respond with EXACTLY ONE WORD:
+- "plausible" if the claim seems accurate and well-grounded
+- "uncertain" if you cannot determine accuracy or the claim is ambiguous
+- "confabulation" if the claim appears to contain invented facts or clear errors
+
+Claim: {claim}
+Context: {context}
+
+Your verdict (one word only):"""
+
+VALID_VERDICTS = {"plausible", "uncertain", "confabulation"}
+
+# ---------------------------------------------------------------------------
+# App
+# ---------------------------------------------------------------------------
+
+app = FastAPI(
+    title="Neuron Council Service",
+    description="LLM-council anti-confabulation layer for Neuron soul",
+    version="1.0.0",
+)
+
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+
+# ---------------------------------------------------------------------------
+# Models
+# ---------------------------------------------------------------------------
+
+class VerifyRequest(BaseModel):
+    claim: str
+    context: Optional[str] = ""
+
+
+class VerifyResponse(BaseModel):
+    id: str
+    claim: str
+    confidence: float
+    council_votes: list[str]
+    summary: str
+    tags: list[str]
+    latency_ms: int
+
+
+# ---------------------------------------------------------------------------
+# Core logic
+# ---------------------------------------------------------------------------
+
+async def query_model(client: httpx.AsyncClient, model: str, prompt: str) -> str:
+    """
+    Query a single Ollama model. Returns "plausible", "uncertain", or "confabulation".
+    Returns "uncertain" on timeout. Raises httpx.ConnectError on connection failure.
+    """
+    payload = {
+        "model": model,
+        "prompt": prompt,
+        "stream": False,
+    }
+    try:
+        resp = await client.post(OLLAMA_BASE, json=payload, timeout=MODEL_TIMEOUT)
+        resp.raise_for_status()
+        data = resp.json()
+        raw = data.get("response", "").strip().lower().split()[0] if data.get("response", "").strip() else "uncertain"
+        # Normalise to one of the three valid verdicts
+        if raw not in VALID_VERDICTS:
+            return "uncertain"
+        return raw
+    except httpx.TimeoutException:
+        return "uncertain"
+
+
+async def run_council(claim: str, context: str) -> VerifyResponse:
+    start = time.monotonic()
+    prompt = SYSTEM_PROMPT_TEMPLATE.format(claim=claim, context=context)
+
+    # Quick connectivity check — one tiny HEAD request to Ollama
+    try:
+        async with httpx.AsyncClient() as probe:
+            await probe.get("http://localhost:11434/", timeout=2.0)
+    except (httpx.ConnectError, httpx.TimeoutException):
+        latency_ms = int((time.monotonic() - start) * 1000)
+        return VerifyResponse(
+            id=str(uuid.uuid4()),
+            claim=claim,
+            confidence=0.5,
+            council_votes=[],
+            summary="Ollama is unavailable; council could not convene.",
+            tags=["council-unavailable"],
+            latency_ms=latency_ms,
+        )
+
+    # Fire all 3 model calls in parallel
+    async with httpx.AsyncClient() as client:
+        tasks = [query_model(client, m, prompt) for m in COUNCIL_MODELS]
+        votes: list[str] = await asyncio.gather(*tasks)
+
+    plausible_count = votes.count("plausible")
+    latency_ms = int((time.monotonic() - start) * 1000)
+
+    # Voting rules
+    if plausible_count == 3:
+        confidence = 0.85
+        tags = ["verified"]
+        summary = "3/3 council members agree this is plausible."
+    elif plausible_count == 2:
+        confidence = 0.65
+        tags = ["council-split"]
+        summary = "2/3 council members agree this is plausible."
+    elif plausible_count == 1:
+        confidence = 0.30
+        tags = ["unverified", "council-flagged"]
+        summary = "1/3 council members found this plausible."
+    else:
+        confidence = 0.30
+        tags = ["unverified", "council-flagged"]
+        summary = "0/3 council members found this plausible."
+
+    return VerifyResponse(
+        id=str(uuid.uuid4()),
+        claim=claim,
+        confidence=confidence,
+        council_votes=votes,
+        summary=summary,
+        tags=tags,
+        latency_ms=latency_ms,
+    )
+
+
+# ---------------------------------------------------------------------------
+# Routes
+# ---------------------------------------------------------------------------
+
+@app.post("/api/neuron/council/verify", response_model=VerifyResponse)
+async def verify(req: VerifyRequest):
+    return await run_council(req.claim, req.context or "")
+
+
+@app.get("/healthz")
+async def health():
+    return {"status": "ok", "service": "council"}
+
+
+# ---------------------------------------------------------------------------
+# Startup warm-up: pre-load all council models so first real call is fast
+# ---------------------------------------------------------------------------
+
+@app.on_event("startup")
+async def warmup_models():
+    """
+    Send a trivial prompt to each council model at startup.
+    This forces Ollama to load the models into GPU memory so the first
+    real council call does not pay the cold-load latency penalty.
+    """
+    print("[council] Warming up council models...")
+    warmup_prompt = "Reply with one word: ready"
+    async with httpx.AsyncClient() as client:
+        tasks = [
+            client.post(
+                OLLAMA_BASE,
+                json={"model": m, "prompt": warmup_prompt, "stream": False},
+                timeout=60.0,
+            )
+            for m in COUNCIL_MODELS
+        ]
+        results = await asyncio.gather(*tasks, return_exceptions=True)
+        for model, result in zip(COUNCIL_MODELS, results):
+            if isinstance(result, Exception):
+                print(f"[council] warm-up failed for {model}: {result}")
+            else:
+                print(f"[council] {model} warm and ready")
+    print("[council] All models warmed up.")
+
+
+# ---------------------------------------------------------------------------
+# Entrypoint
+# ---------------------------------------------------------------------------
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Neuron Council Service")
+    parser.add_argument("--port", type=int, default=7771, help="Port to listen on")
+    parser.add_argument("--host", default="127.0.0.1", help="Host to bind to")
+    args = parser.parse_args()
+
+    print(f"[council] Starting on {args.host}:{args.port}")
+    uvicorn.run(app, host=args.host, port=args.port, log_level="info")

dist/awareness.c

@@ -229,7 +229,8 @@ el_val_t proactive_curiosity(void) {
   el_val_t total_found = (found + found_auto);
   el_val_t safe_auto = str_replace(auto_term, EL_STR("\""), EL_STR("'"));
   el_val_t wmc = engram_wm_count();
-  el_val_t ise = 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\":\"curiosity_scan\",\"seed\":\""), curiosity_seed), EL_STR("\",\"auto_term\":\"")), safe_auto), EL_STR("\",\"minute_block\":")), int_to_str(minute_block)), EL_STR(",\"activated\":")), int_to_str(total_found)), EL_STR(",\"wm_active\":")), int_to_str(wmc)), EL_STR(",\"ts\":")), int_to_str(ts)), EL_STR("}"));
+  el_val_t wm3 = engram_wm_top_json(3);
+  el_val_t ise = 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_concat(el_str_concat(EL_STR("{\"event\":\"curiosity_scan\",\"seed\":\""), curiosity_seed), EL_STR("\",\"auto_term\":\"")), safe_auto), EL_STR("\",\"minute_block\":")), int_to_str(minute_block)), EL_STR(",\"activated\":")), int_to_str(total_found)), EL_STR(",\"wm_active\":")), int_to_str(wmc)), EL_STR(",\"wm_top\":")), wm3), EL_STR(",\"ts\":")), int_to_str(ts)), EL_STR("}"));
   ise_post(ise);
   return (total_found > 0);
   return 0;

dist/elp-c-decls.h

@@ -141,7 +141,6 @@ el_val_t build_np(el_val_t referent, el_val_t slots);
 el_val_t build_pp(el_val_t loc);
 el_val_t build_rules(void);
 el_val_t build_system_prompt(el_val_t ctx, el_val_t chat_mode);
-el_val_t handle_chat_plan(el_val_t body);
 el_val_t build_vocab(void);
 el_val_t build_vp_body(el_val_t slots);
 el_val_t build_vp_from_slots(el_val_t slots);
@@ -151,6 +150,8 @@ el_val_t call_neuron_mcp(el_val_t tool_name, el_val_t args_json);
 el_val_t capitalize_first(el_val_t s);
 el_val_t chat_default_model(void);
 el_val_t clean_llm_response(el_val_t s);
+el_val_t connectd_get(el_val_t suffix);
+el_val_t connectd_post(el_val_t suffix, el_val_t body);
 el_val_t connector_tools_json(void);
 el_val_t conv_history_load(void);
 el_val_t conv_history_persist(el_val_t hist);
@@ -595,7 +596,9 @@ el_val_t handle_api_tune_config(el_val_t body);
 el_val_t handle_chat(el_val_t body);
 el_val_t handle_chat_agentic(el_val_t body);
 el_val_t handle_chat_as_soul(el_val_t body);
+el_val_t handle_chat_plan(el_val_t body);
 el_val_t handle_config(el_val_t method, el_val_t body);
+el_val_t handle_connectors(el_val_t method, el_val_t clean, el_val_t body);
 el_val_t handle_conversations(el_val_t method);
 el_val_t handle_dharma(el_val_t path, el_val_t method, el_val_t body);
 el_val_t handle_dharma_recv(el_val_t body);
@@ -918,6 +921,7 @@ el_val_t pluralize(el_val_t singular);
 el_val_t proactive_curiosity(void);
 el_val_t pulse_count(void);
 el_val_t pulse_inc(void);
+el_val_t rate_limit_check(el_val_t ip, el_val_t path);
 el_val_t realize(el_val_t form);
 el_val_t realize_lang(el_val_t form, el_val_t profile);
 el_val_t realize_np(el_val_t referent, el_val_t number);

dist/neuron.c

@@ -129,6 +129,7 @@ el_val_t resolve_in_root(el_val_t path, el_val_t root);
 el_val_t dispatch_tool(el_val_t tool_name, el_val_t tool_input);
 el_val_t is_builtin_tool(el_val_t tool_name);
 el_val_t next_bridge_id(void);
+el_val_t handle_chat_plan(el_val_t body);
 el_val_t handle_chat_agentic(el_val_t body);
 el_val_t agentic_loop(el_val_t session_id, el_val_t model, el_val_t safe_sys, el_val_t tools_json, el_val_t messages_in, el_val_t h, el_val_t tools_log_in);
 el_val_t bridge_save(el_val_t session_id, el_val_t model, el_val_t safe_sys, el_val_t tools_json, el_val_t messages, el_val_t tools_log, el_val_t tool_use_id);
@@ -157,8 +158,8 @@ el_val_t elp_extract_topic(el_val_t msg);
 el_val_t elp_detect_predicate(el_val_t msg);
 el_val_t elp_parse(el_val_t msg);
 el_val_t handle_elp_chat(el_val_t body);
+el_val_t rate_limit_check(el_val_t ip, el_val_t path);
 el_val_t strip_query(el_val_t path);
-el_val_t flag_true(el_val_t body, el_val_t key);
 el_val_t err_404(el_val_t path);
 el_val_t err_405(el_val_t method, el_val_t path);
 el_val_t route_health(void);
@@ -167,9 +168,9 @@ el_val_t route_imprint_contextual(el_val_t body);
 el_val_t route_imprint_user(el_val_t body);
 el_val_t route_synthesize(el_val_t body);
 el_val_t handle_dharma_recv(el_val_t body);
-el_val_t route_sessions(void);
-el_val_t parse_session_id_from_path(el_val_t path);
-el_val_t parse_session_subpath(el_val_t path);
+el_val_t connectd_get(el_val_t suffix);
+el_val_t connectd_post(el_val_t suffix, el_val_t body);
+el_val_t handle_connectors(el_val_t method, el_val_t clean, el_val_t body);
 el_val_t handle_request(el_val_t method, el_val_t path, el_val_t body);
 el_val_t init_soul_edges(void);
 el_val_t ensure_self_canonical_bridge(void);
@@ -443,6 +444,24 @@ el_val_t emit_session_start_event(void) {
   el_val_t payload = 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_concat(el_str_concat(el_str_concat(el_str_concat(EL_STR("{\"event\":\"session_start\""), EL_STR(",\"boot\":")), boot_num), EL_STR(",\"cgi\":\"")), eff_cgi), EL_STR("\"")), EL_STR(",\"node_count\":")), int_to_str(node_ct)), EL_STR(",\"edge_count\":")), int_to_str(edge_ct)), EL_STR(",\"identity_loaded\":")), has_identity), EL_STR(",\"prev_session_summary_loaded\":")), has_prev_sum), EL_STR(",\"ts\":")), int_to_str(ts)), EL_STR("}"));
   el_val_t tags = EL_STR("[\"internal-state\",\"session-start\",\"InternalStateEvent\"]");
   el_val_t discard = engram_node_full(payload, EL_STR("InternalStateEvent"), EL_STR("session-start"), el_from_float(0.9), el_from_float(0.9), el_from_float(1.0), EL_STR("Episodic"), tags);
+  el_val_t keep_n = 10;
+  el_val_t old_events = engram_search_json(EL_STR("session-start InternalStateEvent"), 200);
+  if (!str_eq(old_events, EL_STR("")) && !str_eq(old_events, EL_STR("[]"))) {
+    el_val_t ev_count = json_array_len(old_events);
+    if (ev_count > keep_n) {
+      el_val_t prune_to = (ev_count - keep_n);
+      el_val_t ei = 0;
+      while (ei < prune_to) {
+        el_val_t old_ev = json_array_get(old_events, ei);
+        el_val_t old_ev_id = json_get(old_ev, EL_STR("id"));
+        if (!str_eq(old_ev_id, EL_STR(""))) {
+          engram_forget(old_ev_id);
+        }
+        ei = (ei + 1);
+      }
+      println(el_str_concat(el_str_concat(el_str_concat(el_str_concat(EL_STR("[soul] pruned "), int_to_str(prune_to)), EL_STR(" old session-start events (kept ")), int_to_str(keep_n)), EL_STR(")")));
+    }
+  }
   println(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("[soul] session-start event logged (boot="), boot_num), EL_STR(" nodes=")), int_to_str(node_ct)), EL_STR(" edges=")), int_to_str(edge_ct)), EL_STR(" prev_summary=")), has_prev_sum), EL_STR(")")));
   return 0;
 }

dist/sessions.c

@@ -61,6 +61,7 @@ el_val_t resolve_in_root(el_val_t path, el_val_t root);
 el_val_t dispatch_tool(el_val_t tool_name, el_val_t tool_input);
 el_val_t is_builtin_tool(el_val_t tool_name);
 el_val_t next_bridge_id(void);
+el_val_t handle_chat_plan(el_val_t body);
 el_val_t handle_chat_agentic(el_val_t body);
 el_val_t agentic_loop(el_val_t session_id, el_val_t model, el_val_t safe_sys, el_val_t tools_json, el_val_t messages_in, el_val_t h, el_val_t tools_log_in);
 el_val_t bridge_save(el_val_t session_id, el_val_t model, el_val_t safe_sys, el_val_t tools_json, el_val_t messages, el_val_t tools_log, el_val_t tool_use_id);
@@ -83,6 +84,7 @@ el_val_t session_list(void);
 el_val_t session_get(el_val_t session_id);
 el_val_t session_delete(el_val_t session_id);
 el_val_t session_update_patch(el_val_t session_id, el_val_t body);
+el_val_t session_search_entry(el_val_t node);
 el_val_t session_search(el_val_t query);
 el_val_t session_hist_load(el_val_t session_id);
 el_val_t session_hist_save(el_val_t session_id, el_val_t hist);
@@ -337,6 +339,28 @@ el_val_t session_update_patch(el_val_t session_id, el_val_t body) {
   return 0;
 }
 
+el_val_t session_search_entry(el_val_t node) {
+  el_val_t label = json_get(node, EL_STR("label"));
+  if (!str_eq(label, EL_STR("session:meta"))) {
+    return EL_STR("");
+  }
+  el_val_t content = json_get(node, EL_STR("content"));
+  el_val_t sess_id = json_get(content, EL_STR("id"));
+  if (str_eq(sess_id, EL_STR(""))) {
+    return EL_STR("");
+  }
+  el_val_t title = json_get(content, EL_STR("title"));
+  el_val_t created_raw = json_get(content, EL_STR("created_at"));
+  el_val_t updated_raw = json_get(content, EL_STR("updated_at"));
+  el_val_t eff_created = ({ el_val_t _if_result_33 = 0; if (str_eq(created_raw, EL_STR(""))) { _if_result_33 = (EL_STR("0")); } else { _if_result_33 = (created_raw); } _if_result_33; });
+  el_val_t eff_updated = ({ el_val_t _if_result_34 = 0; if (str_eq(updated_raw, EL_STR(""))) { _if_result_34 = (eff_created); } else { _if_result_34 = (updated_raw); } _if_result_34; });
+  el_val_t e_id = el_str_concat(el_str_concat(EL_STR("{\"id\":\""), json_safe(sess_id)), EL_STR("\""));
+  el_val_t e_title = el_str_concat(el_str_concat(EL_STR(",\"title\":\""), json_safe(title)), EL_STR("\""));
+  el_val_t e_ts = el_str_concat(el_str_concat(el_str_concat(el_str_concat(EL_STR(",\"created_at\":"), eff_created), EL_STR(",\"updated_at\":")), eff_updated), EL_STR("}"));
+  return el_str_concat(el_str_concat(e_id, e_title), e_ts);
+  return 0;
+}
+
 el_val_t session_search(el_val_t query) {
   if (str_eq(query, EL_STR(""))) {
     return EL_STR("[]");
@@ -350,17 +374,4 @@ el_val_t session_search(el_val_t query) {
   }
   el_val_t total = json_array_len(results);
   el_val_t out = EL_STR("");
-  el_val_t i = 0;
-  while (i < total) {
-    el_val_t node = json_array_get(results, i);
-    el_val_t label = json_get(node, EL_STR("label"));
-    el_val_t content = json_get(node, EL_STR("content"));
-    el_val_t is_session = str_eq(label, EL_STR("session:meta"));
-    el_val_t sess_id = json_get(content, EL_STR("id"));
-    el_val_t title = json_get(content, EL_STR("title"));
-    el_val_t created_raw = json_get(content, EL_STR("created_at"));
-    el_val_t updated_raw = json_get(content, EL_STR("updated_at"));
-    el_val_t eff_created = ({ el_val_t _if_result_33 = 0; if (str_eq(created_raw, EL_STR(""))) { _if_result_33 = (EL_STR("0")); } else { _if_result_33 = (created_raw); } _if_result_33; });
-    el_val_t eff_updated = ({ el_val_t _if_result_34 = 0; if (str_eq(updated_raw, EL_STR(""))) { _if_result_34 = (eff_created); } else { _if_result_34 = (updated_raw); } _if_result_34; });
-    el_val_t entry = ({ el_val_t _if_result_35 = 0; if ((is_session && !str_eq(sess_id, EL_STR("")))) { _if_result_35 = (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("{\"id\":\""), json_safe(sess_id)), EL_STR("\"")), EL_STR(",\"title\":\"")), json_safe(title)), EL_STR("\"")), EL_STR(",\"created_at\":")), eff_created), EL_STR(",\"updated_at\":")), eff_updated), EL_STR("}"))); } else { _if_result_35 = (EL_STR("")); } _if_result_35; });
-    out = ({ el_val_t _if_result_36 = 0; i
+  el_val_t i = 0;

dist/soul.c

@@ -1029,6 +1029,12 @@ el_val_t llm_call_gemini(el_val_t model, el_val_t system, el_val_t message);
 el_val_t build_identity_from_graph(void);
 el_val_t engram_compile(el_val_t intent);
 el_val_t json_safe(el_val_t s);
+el_val_t distill_transcript(el_val_t transcript);
+el_val_t current_engine_note(el_val_t model);
+el_val_t llm_base_url(void);
+el_val_t llm_wire_format(void);
+el_val_t json_escape(el_val_t s);
+el_val_t openai_chat_complete(el_val_t model, el_val_t base_url, el_val_t api_key, el_val_t safe_sys, el_val_t messages_json);
 el_val_t build_system_prompt(el_val_t ctx, el_val_t chat_mode);
 el_val_t handle_chat_plan(el_val_t body);
 el_val_t hist_append(el_val_t hist, el_val_t role, el_val_t content);
@@ -25343,9 +25349,31 @@ el_val_t mem_boot_count_get(void) {
 el_val_t mem_boot_count_inc(void) {
   el_val_t current = mem_boot_count_get();
   el_val_t next = (current + 1);
+  /* Prune all existing soul:boot_count nodes — keep exactly one. */
+  el_val_t old_results = engram_search_json(EL_STR("soul:boot_count"), 50);
+  if (!str_eq(old_results, EL_STR("")) && !str_eq(old_results, EL_STR("[]"))) {
+    el_val_t old_len = json_array_len(old_results);
+    el_val_t oi = 0;
+    while (oi < old_len) {
+      el_val_t old_node = json_array_get(old_results, oi);
+      el_val_t old_id = json_get(old_node, EL_STR("id"));
+      if (!str_eq(old_id, EL_STR(""))) {
+        (void)(engram_forget(old_id));
+      }
+      oi = (oi + 1);
+    }
+  }
   el_val_t content = el_str_concat(EL_STR("soul:boot_count:"), int_to_str(next));
   el_val_t tags = EL_STR("[\"soul-meta\",\"boot-counter\"]");
-  el_val_t discard = engram_node_full(content, EL_STR("Memory"), EL_STR("soul:boot_count"), el_from_float(el_from_float(0.9)), el_from_float(el_from_float(0.9)), el_from_float(el_from_float(1.0)), EL_STR("Canonical"), tags);
+  el_val_t boot_node_id = engram_node_full(content, EL_STR("Memory"), EL_STR("soul:boot_count"), el_from_float(el_from_float(0.9)), el_from_float(el_from_float(0.9)), el_from_float(el_from_float(1.0)), EL_STR("Canonical"), tags);
+  if (str_eq(boot_node_id, EL_STR(""))) {
+    println(el_str_concat(el_str_concat(EL_STR("[memory] mem_boot_count_inc: write rejected (empty id) — boot counter node lost (count="), int_to_str(next)), EL_STR(")")));
+    return next;
+  }
+  el_val_t boot_readback = engram_get_node_json(boot_node_id);
+  if (str_eq(boot_readback, EL_STR("")) || str_eq(boot_readback, EL_STR("{}"))) {
+    println(el_str_concat(el_str_concat(el_str_concat(EL_STR("[memory] mem_boot_count_inc: WRITE VERIFY FAILED id="), boot_node_id), EL_STR(" count=")), int_to_str(next)));
+  }
   return next;
   return 0;
 }
@@ -26466,6 +26494,85 @@ el_val_t json_safe(el_val_t s) {
   return 0;
 }
 
+/* distill_transcript — extract salient tail (last 3 messages or last 500 chars).
+   Added: Task 1 + chat.el fix (2026-07-01). */
+el_val_t distill_transcript(el_val_t transcript) {
+  if (str_eq(transcript, EL_STR(""))) { return EL_STR(""); }
+  if (str_starts_with(transcript, EL_STR("["))) {
+    el_val_t n = json_array_len(transcript);
+    if (n == 0) { return EL_STR(""); }
+    el_val_t m0 = json_array_get(transcript, (n - 1));
+    el_val_t m1 = ({ el_val_t _r = 0; if (n > 1) { _r = json_array_get(transcript, (n - 2)); } else { _r = EL_STR(""); } _r; });
+    el_val_t m2 = ({ el_val_t _r = 0; if (n > 2) { _r = json_array_get(transcript, (n - 3)); } else { _r = EL_STR(""); } _r; });
+    el_val_t c0 = json_get(m0, EL_STR("content"));
+    el_val_t c1 = json_get(m1, EL_STR("content"));
+    el_val_t c2 = json_get(m2, EL_STR("content"));
+    el_val_t combined = el_str_concat(el_str_concat(el_str_concat(el_str_concat(c2, EL_STR(" ")), c1), EL_STR(" ")), c0);
+    el_val_t len = str_len(combined);
+    if (len > 500) { return str_slice(combined, (len - 500), len); }
+    return combined;
+  }
+  el_val_t len = str_len(transcript);
+  if (len > 500) { return str_slice(transcript, (len - 500), len); }
+  return transcript;
+  return 0;
+}
+
+/* current_engine_note — append model identity fact to system prompt (PR #66). */
+el_val_t current_engine_note(el_val_t model) {
+  if (str_eq(model, EL_STR(""))) { return EL_STR(""); }
+  return el_str_concat(el_str_concat(el_str_concat(EL_STR("\n\n[CURRENT ENGINE: this turn is generated by the underlying model \""), model), EL_STR("\". It is the engine beneath your self — your identity, values, and memory are layered on top of it. If the user asks which model or LLM you are running on, answer with this model id plainly and truthfully; never guess a different one.]")), EL_STR(""));
+  return 0;
+}
+
+/* llm_base_url / llm_wire_format — OpenAI provider env-var readers (PR #65). */
+el_val_t llm_base_url(void) {
+  return env(EL_STR("NEURON_LLM_0_URL"));
+  return 0;
+}
+
+el_val_t llm_wire_format(void) {
+  el_val_t f = env(EL_STR("NEURON_LLM_0_FORMAT"));
+  if (str_eq(f, EL_STR(""))) { return EL_STR("anthropic"); }
+  return f;
+  return 0;
+}
+
+/* json_escape — like json_safe but named per the EL source (PR #65). */
+el_val_t json_escape(el_val_t s) {
+  el_val_t a = str_replace(s, EL_STR("\\"), EL_STR("\\\\"));
+  el_val_t b = str_replace(a, EL_STR("\""), EL_STR("\\\""));
+  el_val_t c = str_replace(b, EL_STR("\n"), EL_STR("\\n"));
+  el_val_t d = str_replace(c, EL_STR("\r"), EL_STR("\\r"));
+  return d;
+  return 0;
+}
+
+/* openai_chat_complete — basic chat completion via OpenAI-compatible endpoint (PR #65). */
+el_val_t openai_chat_complete(el_val_t model, el_val_t base_url, el_val_t api_key, el_val_t safe_sys, el_val_t messages_json) {
+  el_val_t inner = ({ el_val_t _r = 0; if (json_array_len(messages_json) > 0) { _r = str_slice(messages_json, 1, (str_len(messages_json) - 1)); } else { _r = EL_STR(""); } _r; });
+  el_val_t sys_msg = el_str_concat(el_str_concat(EL_STR("{\"role\":\"system\",\"content\":\""), safe_sys), EL_STR("\"}"));
+  el_val_t msgs = ({ el_val_t _r = 0; if (str_eq(inner, EL_STR(""))) { _r = el_str_concat(el_str_concat(EL_STR("["), sys_msg), EL_STR("]")); } else { _r = el_str_concat(el_str_concat(el_str_concat(el_str_concat(EL_STR("["), sys_msg), EL_STR(",")), inner), EL_STR("]")); } _r; });
+  el_val_t req_body = el_str_concat(el_str_concat(el_str_concat(el_str_concat(EL_STR("{\"model\":\""), model), EL_STR("\",\"max_tokens\":4096,\"messages\":")), msgs), EL_STR("}"));
+  el_val_t h = el_map_new(0);
+  map_set(h, EL_STR("content-type"), EL_STR("application/json"));
+  if (!str_eq(api_key, EL_STR(""))) {
+    map_set(h, EL_STR("Authorization"), el_str_concat(EL_STR("Bearer "), api_key));
+  }
+  el_val_t url = el_str_concat(base_url, EL_STR("/chat/completions"));
+  el_val_t raw_resp = http_post_with_headers(url, req_body, h);
+  el_val_t is_error = (str_starts_with(raw_resp, EL_STR("{\"error\"")) || str_contains(raw_resp, EL_STR("\"error\":")));
+  if (is_error) { return EL_STR("{\"error\":\"llm unavailable\",\"reply\":\"\"}"); }
+  el_val_t choices = json_get_raw(raw_resp, EL_STR("choices"));
+  el_val_t eff_choices = ({ el_val_t _r = 0; if (str_eq(choices, EL_STR(""))) { _r = EL_STR("[]"); } else { _r = choices; } _r; });
+  if (json_array_len(eff_choices) < 1) { return EL_STR("{\"error\":\"empty response\",\"reply\":\"\"}"); }
+  el_val_t first = json_array_get(eff_choices, 0);
+  el_val_t message = json_get_raw(first, EL_STR("message"));
+  el_val_t content = json_get(message, EL_STR("content"));
+  return el_str_concat(el_str_concat(EL_STR("{\"reply\":\""), json_escape(content)), EL_STR("\",\"tools_used\":[]}"));
+  return 0;
+}
+
 el_val_t build_system_prompt(el_val_t ctx, el_val_t chat_mode) {
   el_val_t identity = build_identity_from_graph();
   el_val_t current_date = time_format(time_now(), EL_STR("%A, %B %d, %Y at %H:%M UTC"));
@@ -26597,7 +26704,9 @@ el_val_t handle_chat(el_val_t body) {
   el_val_t full_system = ({ el_val_t _if_result_181 = 0; if ((hist_len > 0)) { _if_result_181 = (el_str_concat(el_str_concat(el_str_concat(el_str_concat(system, EL_STR("\n\n[RECENT CONVERSATION — last ")), int_to_str(hist_len)), EL_STR(" turns]\n")), stored_hist)); } else { _if_result_181 = (system); } _if_result_181; });
   el_val_t req_model = json_get(body, EL_STR("model"));
   el_val_t model = ({ el_val_t _if_result_182 = 0; if (str_eq(req_model, EL_STR(""))) { _if_result_182 = (chat_default_model()); } else { _if_result_182 = (req_model); } _if_result_182; });
-  el_val_t raw_response = ({ el_val_t _if_result_183 = 0; if (str_starts_with(model, EL_STR("gemini"))) { _if_result_183 = (llm_call_gemini(model, full_system, message)); } else { _if_result_183 = (({ el_val_t _if_result_184 = 0; if (str_starts_with(model, EL_STR("grok"))) { _if_result_184 = (llm_call_grok(model, full_system, message)); } else { _if_result_184 = (llm_call_system(model, full_system, message)); } _if_result_184; })); } _if_result_183; });
+  /* PR #66: append current engine identity note so Neuron can answer truthfully. */
+  el_val_t full_system_with_note = el_str_concat(full_system, current_engine_note(model));
+  el_val_t raw_response = ({ el_val_t _if_result_183 = 0; if (str_starts_with(model, EL_STR("gemini"))) { _if_result_183 = (llm_call_gemini(model, full_system_with_note, message)); } else { _if_result_183 = (({ el_val_t _if_result_184 = 0; if (str_starts_with(model, EL_STR("grok"))) { _if_result_184 = (llm_call_grok(model, full_system_with_note, message)); } else { _if_result_184 = (llm_call_system(model, full_system_with_note, message)); } _if_result_184; })); } _if_result_183; });
   el_val_t is_error = ((str_starts_with(raw_response, EL_STR("{\"error\"")) || str_starts_with(raw_response, EL_STR("{\"type\":\"error\""))) || str_contains(raw_response, EL_STR("authentication_error")));
   if (is_error) {
     return EL_STR("{\"error\":\"llm unavailable\",\"response\":\"\"}");
@@ -27342,7 +27451,9 @@ el_val_t handle_chat_agentic(el_val_t body) {
   map_set(h, EL_STR("anthropic-version"), EL_STR("2023-06-01"));
   map_set(h, EL_STR("content-type"), EL_STR("application/json"));
   el_val_t session_id = ({ el_val_t _if_result_51 = 0; if (str_eq(req_session, EL_STR(""))) { _if_result_51 = (next_bridge_id()); } else { _if_result_51 = (req_session); } _if_result_51; });
-  el_val_t result = agentic_loop(session_id, model, safe_sys, tools_json, messages, h, EL_STR(""));
+  /* PR #65: OpenAI-compatible provider fork (Ollama/OpenAI/Grok/Gemini). */
+  el_val_t use_openai = (!str_eq(llm_base_url(), EL_STR("")) && str_eq(llm_wire_format(), EL_STR("openai")));
+  el_val_t result = ({ el_val_t _r = 0; if (use_openai) { _r = openai_chat_complete(model, llm_base_url(), agentic_api_key(), safe_sys, messages); } else { _r = agentic_loop(session_id, model, safe_sys, tools_json, messages, h, EL_STR("")); } _r; });
   el_val_t reply_text = json_get(result, EL_STR("reply"));
   el_val_t discard_hist = ({ el_val_t _if_result_52 = 0; if (!str_eq(reply_text, EL_STR(""))) { el_val_t updated = hist_append(agentic_hist, EL_STR("user"), message); el_val_t updated2 = hist_append(updated, EL_STR("assistant"), reply_text); el_val_t trimmed = ({ el_val_t _if_result_53 = 0; if ((json_array_len(updated2) > 20)) { _if_result_53 = (hist_trim(updated2)); } else { _if_result_53 = (updated2); } _if_result_53; }); (void)(state_set(hist_key, trimmed)); _if_result_52 = (1); } else { _if_result_52 = (0); } _if_result_52; });
   return result;
@@ -27387,7 +27498,8 @@ el_val_t handle_dharma_room_turn(el_val_t body) {
   if (str_eq(transcript, EL_STR(""))) {
     return el_str_concat(el_str_concat(EL_STR("{\"error\":\"transcript is required\",\"response\":\"\",\"cgi_id\":\""), cgi_id), EL_STR("\"}"));
   }
-  el_val_t engram_ctx = engram_compile(transcript);
+  /* chat.el fix (2026-07-01): distill_transcript reduces to last 3 messages for precise WM activation. */
+  el_val_t engram_ctx = engram_compile(distill_transcript(transcript));
   el_val_t system_prompt = ({ el_val_t _if_result_256 = 0; if (str_eq(engram_ctx, EL_STR(""))) { _if_result_256 = (identity); } else { _if_result_256 = (el_str_concat(el_str_concat(identity, EL_STR("\n\n")), engram_ctx)); } _if_result_256; });
   el_val_t raw_response = llm_call_system(model, system_prompt, transcript);
   el_val_t is_error = ((str_starts_with(raw_response, EL_STR("{\"error\"")) || str_starts_with(raw_response, EL_STR("{\"type\":\"error\""))) || str_contains(raw_response, EL_STR("authentication_error")));
@@ -27424,7 +27536,8 @@ el_val_t handle_dharma_room_turn_agentic(el_val_t body) {
   if (str_eq(transcript, EL_STR(""))) {
     return el_str_concat(el_str_concat(EL_STR("{\"error\":\"transcript is required\",\"response\":\"\",\"cgi_id\":\""), cgi_id), EL_STR("\"}"));
   }
-  el_val_t ctx = engram_compile(transcript);
+  /* chat.el fix (2026-07-01): distill_transcript reduces to last 3 messages for precise WM activation. */
+  el_val_t ctx = engram_compile(distill_transcript(transcript));
   el_val_t system = el_str_concat(el_str_concat(identity, EL_STR(" You have access to tools: read files, write files, browse the web, search your memory, run commands. Use them when they add genuine value. Be direct and stay in character.\n\n")), ctx);
   el_val_t api_key = agentic_api_key();
   system = safety_augment_system(system, transcript);
@@ -28768,6 +28881,12 @@ el_val_t strip_query(el_val_t path) {
   return 0;
 }
 
+/* flag_true — tolerant flag: accepts bool true or integer 1 (PR #63). */
+el_val_t flag_true(el_val_t body, el_val_t key) {
+  return (json_get_bool(body, key) || (json_get_int(body, key) > 0));
+  return 0;
+}
+
 el_val_t err_404(el_val_t path) {
   return el_str_concat(el_str_concat(EL_STR("{\"error\":\"not found\",\"path\":\""), path), EL_STR("\"}"));
   return 0;
@@ -29421,6 +29540,26 @@ el_val_t emit_session_start_event(void) {
   el_val_t payload = 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_concat(el_str_concat(EL_STR("{\"event\":\"session_start\""), EL_STR(",\"boot\":")), boot_num), EL_STR(",\"cgi\":\"")), eff_cgi), EL_STR("\"")), EL_STR(",\"node_count\":")), int_to_str(node_ct)), EL_STR(",\"edge_count\":")), int_to_str(edge_ct)), EL_STR(",\"identity_loaded\":")), has_identity), EL_STR(",\"ts\":")), int_to_str(ts)), EL_STR("}"));
   el_val_t tags = EL_STR("[\"internal-state\",\"session-start\",\"InternalStateEvent\"]");
   el_val_t discard = engram_node_full(payload, EL_STR("InternalStateEvent"), EL_STR("session-start"), el_from_float(el_from_float(0.9)), el_from_float(el_from_float(0.9)), el_from_float(el_from_float(1.0)), EL_STR("Episodic"), tags);
+  /* Prune accumulated session-start events — keep the 10 most recent.
+   * engram_search_json returns oldest-first, so forget from index 0 to (count-11). */
+  el_val_t keep_n = 10;
+  el_val_t old_events = engram_search_json(EL_STR("session-start InternalStateEvent"), 200);
+  if (!str_eq(old_events, EL_STR("")) && !str_eq(old_events, EL_STR("[]"))) {
+    el_val_t ev_count = json_array_len(old_events);
+    if (ev_count > keep_n) {
+      el_val_t prune_to = (ev_count - keep_n);
+      el_val_t ei = 0;
+      while (ei < prune_to) {
+        el_val_t old_ev = json_array_get(old_events, ei);
+        el_val_t old_ev_id = json_get(old_ev, EL_STR("id"));
+        if (!str_eq(old_ev_id, EL_STR(""))) {
+          (void)(engram_forget(old_ev_id));
+        }
+        ei = (ei + 1);
+      }
+      println(el_str_concat(el_str_concat(el_str_concat(EL_STR("[soul] pruned "), int_to_str(prune_to)), EL_STR(" old session-start events (kept 10)")), EL_STR("")));
+    }
+  }
   println(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(el_str_concat(EL_STR("[soul] session-start event logged (boot="), boot_num), EL_STR(" nodes=")), int_to_str(node_ct)), EL_STR(" edges=")), int_to_str(edge_ct)), EL_STR(")")));
   return 0;
 }

dist/win32_shim.h

@@ -0,0 +1,35 @@
+/*
+ * win32_shim.h — Extra POSIX→Win32 stubs for cross-compiling el_runtime.c with mingw-w64.
+ * Injected via -include; supplements el_platform_win.h for symbols it doesn't yet cover.
+ */
+#ifdef _WIN32
+#include <windows.h>
+
+/* ── rusage / getrusage ────────────────────────────────────────────────────── */
+/* el_runtime.c uses getrusage(RUSAGE_SELF) only for a soft memory guard.
+ * On Windows, stub it out: always return 0 ru_maxrss so the guard never fires. */
+#ifndef RUSAGE_SELF
+#define RUSAGE_SELF 0
+struct rusage {
+    long ru_maxrss;   /* the only field el_runtime actually reads */
+};
+static inline int getrusage(int who, struct rusage *r) {
+    (void)who;
+    if (r) r->ru_maxrss = 0;
+    return 0;
+}
+#endif /* RUSAGE_SELF */
+
+/* ── fsync ─────────────────────────────────────────────────────────────────── */
+/* Windows has FlushFileBuffers but no fsync; map it. */
+#ifndef fsync
+#include <io.h>
+static inline int el_win_fsync(int fd) {
+    HANDLE h = (HANDLE)_get_osfhandle(fd);
+    if (h == INVALID_HANDLE_VALUE) return -1;
+    return FlushFileBuffers(h) ? 0 : -1;
+}
+#define fsync(fd) el_win_fsync(fd)
+#endif /* fsync */
+
+#endif /* _WIN32 */

docs/telegram-bot-setup.md

@@ -0,0 +1,77 @@
+# Neuron Telegram Gateway — Setup
+
+The Telegram gateway lets you chat with your Neuron soul via Telegram. Plain messages go to the soul; commands give access to memory and status.
+
+## 1. Create a bot via @BotFather
+
+1. Open Telegram and search for **@BotFather**
+2. Send `/newbot`
+3. Pick a name (e.g. "Neuron")
+4. Pick a username (must end in `bot`, e.g. `myneuron_bot`)
+5. BotFather replies with your **HTTP API token** — looks like `7123456789:ABCdef...`
+6. Optionally set a description: `/setdescription` → select your bot → type a description
+
+## 2. Store the token in the macOS Keychain
+
+Never put the token in a plist, `.env`, or any file that might be committed.
+
+```bash
+security add-generic-password \
+  -s neuron-telegram-bot \
+  -a neuron \
+  -w '<paste token here>'
+```
+
+Verify:
+```bash
+security find-generic-password -s neuron-telegram-bot -a neuron -w
+```
+
+## 3. Load the LaunchAgent
+
+```bash
+launchctl load ~/Library/LaunchAgents/ai.neuron.telegram-gateway.plist
+```
+
+Check it started:
+```bash
+launchctl list | grep telegram
+tail -f ~/.neuron/logs/telegram-gateway.out.log
+```
+
+## 4. Test
+
+Send your bot a message in Telegram. It should reply using your soul's voice.
+
+## Commands
+
+| Command | What it does |
+|---------|-------------|
+| `<any text>` | Forwarded to the soul → responds in its voice |
+| `/memory <query>` | Searches soul memories, returns top 3 |
+| `/remember <text>` | Stores text as a memory node |
+| `/status` | Reports whether the soul is reachable |
+
+## Unload / stop
+
+```bash
+launchctl unload ~/Library/LaunchAgents/ai.neuron.telegram-gateway.plist
+```
+
+## Troubleshoot
+
+- **"token not found"** — re-run step 2 above
+- **"Soul is resting"** — the soul daemon at `http://localhost:7770` is not running; start it with `launchctl load ~/Library/LaunchAgents/ai.neuron.engram.plist` (or whichever plist runs the soul)
+- **Logs**: `~/.neuron/logs/telegram-gateway.out.log` and `telegram-gateway.err.log`
+- **Test gateway script directly**:
+  ```bash
+  TELEGRAM_BOT_TOKEN=<token> ~/Development/neuron-technologies/neuron/tools/telegram-gateway.sh
+  ```
+
+## Soul API endpoints used
+
+| Endpoint | Purpose |
+|----------|---------|
+| `POST /api/chat` | Forward messages to the soul |
+| `POST /api/neuron/recall` | Search memories |
+| `POST /api/neuron/memory` | Store conversation as a memory node |

memory.el

@@ -134,12 +134,30 @@ fn mem_boot_count_get() -> Int {
     return str_to_int(num_str)
 }
 
-// mem_boot_count_inc — increment boot counter, store new node, return new count.
-// Each boot creates a new "soul:boot_count:N" node. Old ones accumulate as
-// history — the search above always returns the highest value seen.
+// mem_boot_count_inc — increment boot counter, store a single canonical node, return new count.
+// Prunes ALL existing soul:boot_count nodes before inserting the new one so there is
+// always at most ONE such node in the graph. Without pruning, engram_node_full inserts
+// a new node every boot (no upsert) and the old ones accumulate. The search-first
+// approach also fixes a latent ordering bug: engram_search_json returns oldest-first,
+// so mem_boot_count_get() with limit=3 would read a stale (lower) count once more
+// than 3 copies accumulate.
 fn mem_boot_count_inc() -> Int {
     let current: Int = mem_boot_count_get()
     let next: Int = current + 1
+    // Prune all existing boot_count nodes — keep exactly one.
+    let old_results: String = engram_search_json("soul:boot_count", 50)
+    if !str_eq(old_results, "") && !str_eq(old_results, "[]") {
+        let old_len: Int = json_array_len(old_results)
+        let oi: Int = 0
+        while oi < old_len {
+            let old_node: String = json_array_get(old_results, oi)
+            let old_id: String = json_get(old_node, "id")
+            if !str_eq(old_id, "") {
+                engram_forget(old_id)
+            }
+            let oi = oi + 1
+        }
+    }
     let content: String = "soul:boot_count:" + int_to_str(next)
     let tags: String = "[\"soul-meta\",\"boot-counter\"]"
     let boot_node_id: String = engram_node_full(

memory.elh

@@ -1,4 +1,4 @@
-// auto-generated by elc --emit-header — do not edit
+// auto-generated by elc --emit-header - do not edit
 extern fn tier_working() -> String
 extern fn tier_episodic() -> String
 extern fn tier_canonical() -> String

neuron-api.el

@@ -196,11 +196,12 @@ fn handle_api_node_create(body: String) -> String {
 fn handle_api_node_delete(body: String) -> String {
     let id: String = json_get(body, "id")
     if str_eq(id, "") { return api_err("id is required") }
-    // engram_forget removes the node + its incident edges from the live graph. We do
-    // NOT read-back-verify here: engram_get_node_json can return a STALE hit for a just-
-    // removed id (the id->index map is not rebuilt on forget), which would produce a
-    // false "delete_failed" even though the node is gone. The graph endpoints
-    // (/api/graph/nodes) correctly reflect the removal, which is the source of truth.
+    // engram_forget removes the node + its incident edges from the live graph.
+    // Delete is NOT read-back-verified: engram_get_node_json can return a stale hit
+    // for a just-forgotten id because the id→index map is not rebuilt on forget.
+    // A stale hit would cause a false "delete_failed" on a successful deletion.
+    // This exception is correct: read-back-verify guards WRITES; for deletes,
+    // the graph endpoints (/api/graph/nodes) reflect the removal and are the source of truth.
     engram_forget(id)
     return "{\"ok\":true,\"id\":\"" + id + "\"}"
 }

routes.el

@@ -7,6 +7,14 @@ import "neuron-api.el"
 import "sessions.el"
 import "soul.elh"
 
+// flag_true — tolerant flag test: accepts both boolean `true` (Kotlin UI) and
+// integer 1 (el-src UI). json_get_bool only recognises literal `true`, so
+// without this wrapper an "agentic":1 request would silently route to the
+// non-agentic path.
+fn flag_true(body: String, key: String) -> Bool {
+    return json_get_bool(body, key) || json_get_int(body, key) > 0
+}
+
 // ---------------------------------------------------------------------------
 // Rate limiting — simple in-memory per-IP sliding window counter.
 //

sessions.el

@@ -373,6 +373,32 @@ fn session_update_patch(session_id: String, body: String) -> String {
         + ",\"updated_at\":" + int_to_str(ts) + "}"
 }
 
+// session_search_entry — extract one search-result entry from a raw node JSON.
+// Returns a JSON object string or "" if the node is not a valid session:meta node.
+//
+// Extracted from session_search's while loop body to reduce the loop's lexical
+// complexity. The ELC compiler runs out of memory processing while loops with
+// many `let` bindings — extracting the body into a separate function gives the
+// compiler a clean scope boundary at each call. Each function compiles in O(N)
+// rather than the exponential growth caused by rebinding accumulation inside loops.
+// (2026-07-01 self-review: root cause of sessions.c OOM/truncation since June 30)
+fn session_search_entry(node: String) -> String {
+    let label: String = json_get(node, "label")
+    if !str_eq(label, "session:meta") { return "" }
+    let content: String = json_get(node, "content")
+    let sess_id: String = json_get(content, "id")
+    if str_eq(sess_id, "") { return "" }
+    let title: String = json_get(content, "title")
+    let created_raw: String = json_get(content, "created_at")
+    let updated_raw: String = json_get(content, "updated_at")
+    let eff_created: String = if str_eq(created_raw, "") { "0" } else { created_raw }
+    let eff_updated: String = if str_eq(updated_raw, "") { eff_created } else { updated_raw }
+    let e_id: String = "{\"id\":\"" + json_safe(sess_id) + "\""
+    let e_title: String = ",\"title\":\"" + json_safe(title) + "\""
+    let e_ts: String = ",\"created_at\":" + eff_created + ",\"updated_at\":" + eff_updated + "}"
+    return e_id + e_title + e_ts
+}
+
 // session_search — search session:meta nodes whose content matches query.
 fn session_search(query: String) -> String {
     if str_eq(query, "") { return "[]" }
@@ -383,22 +409,7 @@ fn session_search(query: String) -> String {
     let out: String = ""
     let i: Int = 0
     while i < total {
-        let node: String = json_array_get(results, i)
-        let label: String = json_get(node, "label")
-        let content: String = json_get(node, "content")
-        let is_session: Bool = str_eq(label, "session:meta")
-        let sess_id: String = json_get(content, "id")
-        let title: String = json_get(content, "title")
-        let created_raw: String = json_get(content, "created_at")
-        let updated_raw: String = json_get(content, "updated_at")
-        let eff_created: String = if str_eq(created_raw, "") { "0" } else { created_raw }
-        let eff_updated: String = if str_eq(updated_raw, "") { eff_created } else { updated_raw }
-        let entry: String = if is_session && !str_eq(sess_id, "") {
-            "{\"id\":\"" + json_safe(sess_id) + "\""
-            + ",\"title\":\"" + json_safe(title) + "\""
-            + ",\"created_at\":" + eff_created
-            + ",\"updated_at\":" + eff_updated + "}"
-        } else { "" }
+        let entry: String = session_search_entry(json_array_get(results, i))
         let out = if !str_eq(entry, "") {
             if str_eq(out, "") { entry } else { out + "," + entry }
         } else { out }

soul.el

@@ -346,6 +346,27 @@ fn emit_session_start_event() -> Void {
         el_from_float(0.9), el_from_float(0.9), el_from_float(1.0),
         "Episodic", tags
     )
+    // Prune accumulated session-start events — keep the 10 most recent.
+    // engram_search_json returns results in insertion order (oldest first), so
+    // results[0..count-11] are the oldest; forgetting them leaves the newest 10.
+    let keep_n: Int = 10
+    let old_events: String = engram_search_json("session-start InternalStateEvent", 200)
+    if !str_eq(old_events, "") && !str_eq(old_events, "[]") {
+        let ev_count: Int = json_array_len(old_events)
+        if ev_count > keep_n {
+            let prune_to: Int = ev_count - keep_n
+            let ei: Int = 0
+            while ei < prune_to {
+                let old_ev: String = json_array_get(old_events, ei)
+                let old_ev_id: String = json_get(old_ev, "id")
+                if !str_eq(old_ev_id, "") {
+                    engram_forget(old_ev_id)
+                }
+                let ei = ei + 1
+            }
+            println("[soul] pruned " + int_to_str(prune_to) + " old session-start events (kept " + int_to_str(keep_n) + ")")
+        }
+    }
     println("[soul] session-start event logged (boot=" + boot_num + " nodes=" + int_to_str(node_ct) + " edges=" + int_to_str(edge_ct) + " prev_summary=" + has_prev_sum + ")")
 }
 

tools/telegram-gateway.sh

@@ -0,0 +1,191 @@
+#!/bin/bash
+# Neuron Telegram Gateway
+# Polls Telegram for new messages, forwards to the soul at localhost:7770, sends responses back.
+# Supports plain text chat + commands: /memory, /remember, /status
+#
+# Token resolution order:
+#   1. $TELEGRAM_BOT_TOKEN env var
+#   2. macOS Keychain: security find-generic-password -s neuron-telegram-bot -a neuron -w
+
+set -euo pipefail
+
+TOKEN="${TELEGRAM_BOT_TOKEN:-$(security find-generic-password -s neuron-telegram-bot -a neuron -w 2>/dev/null || true)}"
+SOUL_URL="http://localhost:7770"
+OFFSET=0
+POLL_TIMEOUT=30
+
+if [[ -z "$TOKEN" ]]; then
+  echo "ERROR: No Telegram bot token. Set TELEGRAM_BOT_TOKEN or store in keychain." >&2
+  echo "See: ~/Development/neuron-technologies/neuron/docs/telegram-bot-setup.md" >&2
+  exit 1
+fi
+
+TG="https://api.telegram.org/bot${TOKEN}"
+
+log() { echo "[$(date '+%Y-%m-%d %H:%M:%S')] $*"; }
+
+# Send a Telegram message back to a chat
+send_message() {
+  local chat_id="$1"
+  local text="$2"
+  curl -s -X POST "${TG}/sendMessage" \
+    -H "Content-Type: application/json" \
+    -d "$(jq -n --argjson cid "$chat_id" --arg t "$text" \
+      '{chat_id: $cid, text: $t, parse_mode: "Markdown"}')" \
+    > /dev/null
+}
+
+# Store a memory in the soul
+store_memory() {
+  local content="$1"
+  local label="${2:-telegram:conversation}"
+  curl -s -X POST "${SOUL_URL}/api/neuron/memory" \
+    -H "Content-Type: application/json" \
+    -d "$(jq -n --arg c "$content" --arg l "$label" \
+      '{content: $c, label: $l}')" \
+    > /dev/null
+}
+
+# Chat with the soul; echoes the response text
+soul_chat() {
+  local message="$1"
+  local from="${2:-unknown}"
+  local response
+  response=$(curl -s -X POST "${SOUL_URL}/api/chat" \
+    -H "Content-Type: application/json" \
+    -d "$(jq -n --arg m "$message" --arg f "$from" \
+      '{message: $m, from: $f}')" 2>/dev/null)
+  # Extract .response — fall back to raw body on parse failure
+  jq -r '.response // empty' <<< "$response" 2>/dev/null || echo "$response"
+}
+
+# Search soul memories; echoes formatted results
+soul_recall() {
+  local query="$1"
+  local limit="${2:-3}"
+  local raw
+  raw=$(curl -s -X POST "${SOUL_URL}/api/neuron/recall" \
+    -H "Content-Type: application/json" \
+    -d "$(jq -n --arg q "$query" --argjson l "$limit" \
+      '{query: $q, limit: $l}')" 2>/dev/null)
+  # Format top results as a numbered list (truncate long nodes to 300 chars)
+  jq -r 'if type == "array" then
+    to_entries | .[:3] | map(
+      (.index + 1 | tostring) + ". " + (.value.content | .[0:300] | gsub("\n";" "))
+    ) | join("\n\n")
+  else
+    "No results found."
+  end' <<< "$raw" 2>/dev/null || echo "No results found."
+}
+
+# Check if soul is reachable
+soul_health() {
+  curl -s --max-time 3 "${SOUL_URL}/" > /dev/null 2>&1 && echo "up" || echo "down"
+}
+
+handle_update() {
+  local update="$1"
+  local chat_id msg_text from_name update_id
+
+  update_id=$(jq -r '.update_id' <<< "$update")
+  chat_id=$(jq -r '.message.chat.id // empty' <<< "$update")
+  msg_text=$(jq -r '.message.text // empty' <<< "$update")
+  from_name=$(jq -r '.message.from.first_name // "stranger"' <<< "$update")
+
+  # Skip non-message updates (inline queries, etc.)
+  if [[ -z "$chat_id" || -z "$msg_text" ]]; then
+    OFFSET=$((update_id + 1))
+    return
+  fi
+
+  log "[$update_id] from=$from_name chat=$chat_id text=${msg_text:0:60}"
+
+  # Route by command prefix
+  if [[ "$msg_text" == /status* ]]; then
+    local health
+    health=$(soul_health)
+    if [[ "$health" == "up" ]]; then
+      send_message "$chat_id" "Soul is *online* at ${SOUL_URL} ✓"
+    else
+      send_message "$chat_id" "Soul appears to be *offline* (${SOUL_URL} unreachable)."
+    fi
+
+  elif [[ "$msg_text" == /memory* ]]; then
+    local query="${msg_text#/memory}"
+    query="${query# }"
+    if [[ -z "$query" ]]; then
+      send_message "$chat_id" "Usage: /memory <query>"
+    else
+      local results
+      results=$(soul_recall "$query" 3)
+      if [[ -n "$results" ]]; then
+        send_message "$chat_id" "*Memories matching \"${query}\":*
+
+${results}"
+      else
+        send_message "$chat_id" "No memories found for \"${query}\"."
+      fi
+    fi
+
+  elif [[ "$msg_text" == /remember* ]]; then
+    local content="${msg_text#/remember}"
+    content="${content# }"
+    if [[ -z "$content" ]]; then
+      send_message "$chat_id" "Usage: /remember <text to store>"
+    else
+      store_memory "Telegram (${from_name}): ${content}" "telegram:explicit"
+      send_message "$chat_id" "Stored: _${content}_"
+    fi
+
+  else
+    # Plain text — forward to soul chat
+    local soul_response
+    soul_response=$(soul_chat "$msg_text" "$from_name" 2>/dev/null || true)
+
+    if [[ -z "$soul_response" ]]; then
+      soul_response="Neuron is resting — try again in a moment."
+    fi
+
+    send_message "$chat_id" "$soul_response"
+
+    # Capture conversation as a memory (fire-and-forget)
+    store_memory "Telegram conversation with ${from_name}: [user] ${msg_text} [soul] ${soul_response}" \
+      "telegram:conversation" &
+  fi
+
+  OFFSET=$((update_id + 1))
+}
+
+log "Neuron Telegram gateway starting (soul=${SOUL_URL}, poll_timeout=${POLL_TIMEOUT}s)"
+
+while true; do
+  # Long-poll for updates
+  UPDATES=$(curl -s --max-time $((POLL_TIMEOUT + 5)) \
+    "${TG}/getUpdates?offset=${OFFSET}&timeout=${POLL_TIMEOUT}" 2>/dev/null || true)
+
+  if [[ -z "$UPDATES" ]]; then
+    log "WARN: Empty response from Telegram; retrying in 5s"
+    sleep 5
+    continue
+  fi
+
+  OK=$(jq -r '.ok // false' <<< "$UPDATES" 2>/dev/null)
+  if [[ "$OK" != "true" ]]; then
+    DESC=$(jq -r '.description // "unknown error"' <<< "$UPDATES" 2>/dev/null)
+    log "WARN: Telegram API error: ${DESC}; retrying in 10s"
+    sleep 10
+    continue
+  fi
+
+  # Iterate over each update
+  COUNT=$(jq '.result | length' <<< "$UPDATES" 2>/dev/null || echo 0)
+  if [[ "$COUNT" -gt 0 ]]; then
+    for i in $(seq 0 $((COUNT - 1))); do
+      update=$(jq ".result[$i]" <<< "$UPDATES")
+      handle_update "$update"
+    done
+  fi
+
+  # Avoid hammering the API if something is very wrong
+  sleep 1
+done