The approve endpoint was permanently broken for all sessions going through
the modern agentic_loop path. agentic_loop suspends via bridge_save() into
mcp_bridge:<session_id>, but handle_session_approve was reading from
pending_tool_<session_id> — a different key — so it always returned
"no pending tool for session".
Replace the body of handle_session_approve with a two-path design:
Modern path: check mcp_bridge:<session_id> first. If the blob is there,
dispatch_tool() on allow (or build the denial string), then delegate to
agentic_resume() which re-enters agentic_loop from the exact suspension
point. This is the path all live sessions take.
Legacy path: if only pending_tool_<session_id> exists (in-flight session
from before this deploy), synthesise a bridge blob from the stored
messages_so_far and route through agentic_resume() as well. The stale
inline agentic loop (90 lines, agentic_tools_literal only, no MCP
connector support, no bridge suspension) is removed entirely.
routes.el already calls handle_session_approve correctly — no change needed.
- sessions.el: new sessions module with session management and approval gate
- routes.el: wire /api/sessions routes (list, get, create, approve, tool_result)
- chat.el: thread-aware activation — short messages anchor to last reply
before engram compilation so follow-ups stay on-topic
- chat.el: agentic path tracks per-session history (session_hist_{id})
instead of shared conv_history, seeding each turn with prior context
- chat.el: add call_neuron_mcp, dispatch_tool, is_builtin_tool, next_bridge_id
agentic_loop, bridge_save, agentic_resume, handle_tool_result
- dist/soul: rebuild with all of the above
Short/ambiguous messages (< 50 chars) now use the last reply as the
engram activation seed instead of the bare message. Prevents strong
off-topic memory nodes from hijacking replies when the user is clearly
continuing an existing thread.
Also gives handle_chat_agentic session continuity: reads/writes history
keyed by session_id (falling back to global conv_history), seeds the
LLM messages array with prior turns, and saves replies back so the
next turn has context.
Applies connector-specific additions from feat/connectors-soul:
- chat.el: connector_tools_json(), agentic_tools_all(), call_mcp_bridge(),
tool_auto_approved() and mcp__ dispatch in dispatch_tool()
- routes.el: connectd_get/post, handle_connectors(), /api/connectors routing
in GET and POST sections
- MEMORY_RECALL_BUG.md: investigation notes on memory retrieval failure
The agentic loop rewrite in the source branch was not applied — it conflicts
with the tool-bridge pattern from PR #5 which is the chosen design for
client-side MCP tool execution. The connectors themselves are now fully
wired: connector tools surface as mcp__<server>__<tool> in the tools array
and dispatch to neuron-connectd via call_mcp_bridge().
Resolves conflicts by keeping main's full safety/stewardship/imprint implementations.
PR #9 uniquely contributes: layered_cycle() in soul.el, route wiring in routes.el,
soul.elh export, and the layer composition test suite.
dist/*.c and *.elh are elc transpiler output. CI's header-gen step still
greps dist/*.c, so they stay tracked, but a single soul change regenerates
~57k lines of dist/neuron.c + dist/soul.c that bury the real source diff and
poison both human and agent PR review. Mark them -diff + linguist-generated so
PRs show only the real changes. Build pipeline unchanged.
Linux elb generates individual .c files; soul.c does not contain merged
imports (unlike macOS elb which produces a unified file). Re-link all
dist/*.c manually with soul.c listed first so its real main() wins, and
--allow-multiple-definition to silence GNU ld's duplicate symbol errors.
All duplicates are identical (same El source, different compile units).
The El compiler inlines imported modules into each module's .c file.
On macOS, ld64 accepts duplicate strong symbols silently. On Linux,
GNU ld rejects them. soul.c is a fully merged file — every function
from every imported module is present in it — so linking only soul.c
against el_runtime.c produces a correct binary with no duplicates.
- Add stub implementations of safety.el, stewardship.el, and imprint.el
with their .elh headers so the branch compiles without the dependency
branches (feat/layer-safety, feat/layer-stewardship, feat/layer-imprint).
Each stub documents the layer contract it must satisfy when replaced.
- Fix GET /api/chat bypass: update the GET branch in handle_request to
call layered_cycle() consistently with the POST branch, rather than
calling handle_chat() directly and skipping the consciousness stack.
- Export layered_cycle() from soul.elh (and dist/soul.elh) so routes.el
can resolve the symbol via the header import.
- Fix steward_action else branch: add explicit handling for "block"
(returns safe refusal immediately, skips L3) and "redirect" (uses
redirect_to field). Unknown actions now log a warning and fall back to
the screened input rather than silently passing an empty string to
imprint_respond().
- Document hard_bell path: clarify that omitting auto_persist/history
update is intentional security isolation, and document the safety_validate
second-param sentinel contract ("hard_bell" vs screen_action).
- steward_log_event (line 14): add println after let discard so the
function's last expression is Void, fixing the type mismatch on a
Void-declared function
- steward_get_mission (lines 40-43): remove non-Config fallthrough that
allowed any Episodic/Working node to silently override the mission;
only Config nodes are now authoritative
- steward_align signal_deceive (line 56): widen 'deceive the user' to
'deceive' to catch variants like 'deceive users', 'deceive them', etc.
- steward_align signal_hide (line 57): tighten 'hide from' to
'hide from the user' to eliminate false positives on legitimate inputs
like 'hide from a background process' or 'hide from view'
- stewardship.elh: document that steward_log_event is an internal helper
exported only because El has no access modifiers; callers should not
invoke it directly
35 test cases covering all five public functions:
steward_align (pass-through, all five misalignment signals, empty input,
json_get field extraction, redirect shape), steward_validate_imprint
(standard tools, platform-only tools with/without platform_auth,
auth=false string), steward_cgi_check (all four gated actions, non-gated
actions, empty action, action name echoed in response), and
steward_get_mission (non-empty, contains "integrity", not an error object).
Also documents the known bug surface from the code review: the &&
operator in steward_get_mission and the non-Config fallthrough — tests
are written against the actual runtime behaviour so they will catch
regressions when those bugs are fixed.
Covers: imprint_current base fallback, unload idempotency, load miss →
ok=false, ok field presence, respond passthrough for base/empty/unknown
IDs, graceful fallback after unload, surface_knowledge and
surface_memory_read return-type guarantees, base-scoped knowledge
equality, no-annotation invariant for base, empty-ID load rejection, and
failed-load state immutability.
Syntax follows El constraints: no Bool annotations, no &&/||, no unary !.
proactive_curiosity() now uses the top working-memory node's first label
word as a 4th activation seed alongside the 4 rotating fixed sets. This
breaks deterministic exploration that was reinforcing the same subgraph
every cycle and creates a self-referencing loop: curiosity radiates from
whatever is most salient right now, mirroring the brain's default-mode-
network resting-state dynamics. str_find_chars on " :([" extracts the
first meaningful word; sp > 3 guards against bracket-prefixed labels.
auto_term field added to curiosity_scan ISE for observability.
When handle_chat_agentic hits a tool the soul cannot run in-process (an MCP
connector/plugin surfaced by the Kotlin desktop app), instead of returning
"unknown tool" it now suspends the agentic loop and returns a tool_pending
envelope so the CLIENT executes the tool and posts the result back. Built-in
tools (read_file/write_file/web_get/search_memory/run_command) and Anthropic's
native web_search are unchanged.
Client contract:
- Soul returns (HTTP 200) on an unknown tool:
{ "tool_pending": true, "session_id": "br-...", "call_id": "<tool_use_id>",
"tool_name": "...", "tool_input": { ... }, "model": "...",
"agentic": true, "tools_used": [...] }
- Client runs the MCP tool, then POSTs to
/api/sessions/{session_id}/tool_result
with body:
{ "call_id": "<the call_id from the envelope>",
"content": "<MCP tool output as a string>" }
- Soul resumes the loop and returns the same envelope shape: either a final
{ "reply": ..., "tools_used": [...] }
or another tool_pending if the continuation needs a further MCP tool
(fully chainable). Saved continuation is one-shot (cleared on resume).
elc-verified (--target=c, exit 0, no stderr) on chat.el, routes.el, and the
full soul.el import graph. Needs Will's build to ship.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The UI needs full memory CRUD; the soul had create (handle_api_remember)
and the older /memory/forget + /memory/evolve, but no endpoints matching
the UI's delete/update contract.
POST /api/neuron/memory/delete {"id"}
Hard delete. engram_forget is a true delete primitive (removes the node
and all incident edges from the engram store), so no soft-delete
fallback is needed. Unlike /memory/forget, this checks the node exists
first - engram_forget silently no-ops on unknown ids, and a bad id must
return an error, not fake success. Protected identity/values nodes are
blocked, same as the other accumulation-path handlers.
POST /api/neuron/memory/update {"id","content"}
Evolve-style update. The engram runtime has no in-place node mutation
primitive (only node-create, strengthen, forget, connect), so update
creates a new Memory node and wires a supersedes edge to the prior one,
same pattern as handle_api_evolve_knowledge. Unlike /memory/evolve, id
is required and must reference an existing node; create+link delegates
to handle_api_evolve_memory. Returns {id, supersedes, ok}.
Both files syntax-checked with elc --target=c (exit 0, no stderr).
Compile-verified only - local builds cannot run the soul; needs Will's
build for runtime verification.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Two fixes:
1. ise_post was only escaping " in content strings. When wm_top contained
node labels with \n (backslash-n escape sequences from jb_emit_escaped),
the HTTP Engram server's JSON parser decoded \n as a literal newline in
the stored content, making heartbeat ISEs unparseable. Fix: escape
backslashes first, then quotes, then \n and \r — matching make_action's
existing pattern. Result: heartbeat ISEs now parse cleanly.
2. Soul daemon (dist/neuron) was missing — the build command in the prompt
was linking all 46 dist/*.c files together, causing 1092 duplicate symbol
errors. EL compiles transitive imports inline so neuron.c is self-contained;
correct build links ONLY neuron.c + el_runtime.c. Daemon now starts.
handle_chat_agentic now always attaches Anthropic's native web_search_20250305
tool instead of gating it behind a per-request web_search flag. Web search is a
built-in capability: the model invokes it only when a query needs fresh info
(max_uses:5 caps it), so there is no user-facing toggle. The body's web_search
field is now ignored (back-compat — old UI clients sending it cause no harm).
Pairs with neuron-ui removing the chat-input web search toggle.
Note: .el change only — no elc on the authoring machine; reviewer builds/verifies.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Tooling built on Tim's machine to run Neuron from the terminal as a
Claude Code session (identity + graph memory + agency) instead of
relaying to the soul's /api/chat.
- cli/neuron_recall.py BM25 read over the engram snapshot + CLI memories
(works around pinned-only soul search)
- cli/neuron_remember.py reliable local memory writes with read-back verify
(works around the corrupting capture endpoint)
- cli/neuron-chat.py standalone direct-chat REPL with per-turn memory injection
- cli/neuron_mcp.py stdlib MCP server (chat/search) with graceful degradation
- cli/CLAUDE.md.example the operating identity that makes Claude Code run as Neuron
- cli/HANDOFF.md soul-side bugs to fix so this becomes unnecessary
Scaffolding/proposal - intended to be retired once the soul does native
retrieval, correct persistence, and a real CLI identity/voice surface.
Pairs with the runtime model-passthrough + UTF-8 fixes in the el repo.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
will.anderson
Tim Lingo