feat: Wave 0-3 — EA/EFM kernel overhaul#1
Merged
chazmaniandinkle merged 10 commits intocogos-dev:mainfrom Apr 7, 2026
Merged
Conversation
…ction - evictForBudgetMode now evicts user+assistant pairs atomically (prevents orphaned assistant answers without their triggering prompt) - Added estTokensPrecise() with rune-aware heuristics for CJK, JSON, and code. Activated when iris pressure > 0.8 via WithIrisSignal option. - Replaced extractGoal prefix matching with keyword-anywhere matching across action/understand/operate categories. Handles polite prefixes. All three bugs identified during architectural review session. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-Authored-By: Codex gpt-5.4 <noreply@openai.com>
- Added model profile system to OllamaProvider with known profiles for gemma4:e4b, gemma4:e2b, qwen3.5:9b - New CapToolCallValidation capability — Gemma 4 advertises this but NOT CapToolUse, enabling the router to route tool-heavy tasks to providers with reliable tool support - Added local_model config field (default: gemma4:e4b) - Router wired to honor local_model override with provider rejection tracking Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-Authored-By: Codex gpt-5.4 <noreply@openai.com>
- ValidateToolCall rejects: unknown tools, missing required params, type mismatches, and embedded result fields (hallucination detection) - Rejected calls fed back to model as system message for retry - Proprioceptive logging of rejections for shadow model training - Per-provider rejection counter feeds router sovereignty gradient - ToolCallValidationEnabled config (default true) - Trusted providers with CapToolUse skip validation automatically Addresses known Gemma 4 tool-call fabrication issue. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-Authored-By: Codex gpt-5.4 <noreply@openai.com>
… adapters - StreamTailer interface + TailerManager for lifecycle management - Zero-dependency FileWatcher with polling, truncation, and rotation handling - ClaudeCodeTailer normalizes ~/.claude/ JSONL logs to CogBlocks - OpenClawTailer normalizes OpenClaw JSONL with directory mode (auto-discovers files) - Both set proper provenance: source_channel, normalized_by The digestion pipeline converts external harness logs into cog-native events. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-Authored-By: Codex gpt-5.4 <noreply@openai.com>
- ConsolidationAction reads recent ledger events, groups by session, extracts keyword topics, writes summary CogBlocks - Wired into process.go heartbeat path during Dormant state (1hr default) - Archive emission after consolidation records consolidated session_ids - ArchivedSessions() helper for querying which sessions are consolidated - Archive events filtered from future consolidation passes (no feedback loops) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-Authored-By: Codex gpt-5.4 <noreply@openai.com>
- SyncWatcher polls .cog/sync/inbox/ for BEP-synced SyncEnvelopes with structural validation (version, timestamp, hash format, kind) - ConstellationBridge interface: EmitHeartbeat, TrustSnapshot, Start/Stop - NilBridge for standalone mode (healthy defaults, zero overhead) - Wired into process.go: emitHeartbeat() calls bridge with full state payload (process state, field size, coherence/nucleus fingerprints, ledger head) - Bridge receipt hash attached as CogBlock artifact Per the constellation-kernel bridge specification (D1). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-Authored-By: Codex gpt-5.4 <noreply@openai.com>
- Lead with "externalized attention and executive function modulation"
- Added "The model generates. The substrate thinks." framing
- Expanded ecosystem table (5 repos with verb-based descriptions)
- Reframed mod3 as modality bus ("how it ACTS")
- Updated status to reflect Wave 0+1 implementation work
- Added EA/EFM section to Core Ideas
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- TailerManager initialized during Run() from DigestPaths config - Select-loop case ingests CogBlocks during Receptive/Active states - Drops ingestion during Consolidating/Dormant (state-gated) - DigestPaths config: map[string]string of adapter_name -> path (default empty) - Integration test with mock tailer verifies ledger write Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-Authored-By: Codex gpt-5.3-codex <noreply@openai.com>
- Tool loop: wrong-param-type rejection, rejection counter tracking, trusted-provider skip parallelized - Constellation bridge: full payload through NilBridge.EmitHeartbeat - Sync watcher: non-JSON ignore, empty directory handling - Claude Code tailer: tool_use/tool_result kind normalization, malformed line safety - OpenClaw tailer: directory-mode discovery, malformed JSON skipping Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-Authored-By: Codex gpt-5.3-codex <noreply@openai.com>
Five test scenarios covering: 1. Standalone kernel (health, chat, ledger, context, shutdown) 2. Gemma 4 integration (local inference, tool-call gate, proprioceptive log) 3. Digestion pipeline (JSONL tailing, CogBlock ingestion, state gating) 4. Constellation bridge (heartbeat payload, SyncEnvelope detection) 5. Full stack (kernel + mod3 + constellation cross-system communication) Spec document — no test code. Aligned with existing e2e-test.sh patterns. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-Authored-By: Codex gpt-5.3-codex <noreply@openai.com>
chazmaniandinkle
referenced
this pull request
in chazmaniandinkle/cogos
Apr 14, 2026
Foundation layer for the foveated context rendering pipeline: - cogdoc_service.go: CogDocService with WriteAndSync/PatchAndSync — single write-through path ensuring index refresh + field boost + ledger emit on every CogDoc mutation (fixes Codex finding #1) - context_frame.go: ContextFrame type with named blocks (nucleus, project, knowledge, node, field, events, focus), tier priorities, stability hints for KV cache, budget-aware eviction, and HTML comment rendering - mcp_server.go: 3 MCP Resources (cogos://state, cogos://nucleus, cogos://field) alongside the 10 tools — proper tool/resource separation per MCP spec - uri_resolve_test.go: 15 tests covering URI holographic pointer round-trips across all forms (cog://, cog:, relative, absolute) Gate G0→1: go build + go test both pass clean. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This was referenced Apr 21, 2026
chazmaniandinkle
added a commit
that referenced
this pull request
Apr 21, 2026
Track 1 of the Windows dev-preview rollout (PRs #1 and #2 in Agent K's release audit). Makefile: - Add windows-amd64 and windows-arm64 cross-compile targets, mirroring release.yml exactly (CGO_ENABLED=0, ./cmd/cogos/ entry point, .exe suffix). Extends the existing PLATFORMS list rather than introducing a new pattern. docs/RELEASING.md: - Add a "Installing on Windows (developer preview)" section covering PowerShell download, SmartScreen "More info -> Run anyway" workaround for the unsigned binary, %LOCALAPPDATA%\cogos\ install path with User PATH update, and a version/serve/health sanity check using the actual subcommands exposed by internal/engine/cli.go. - Note that Windows Service / SCM integration is a follow-up, not part of v0.x. No version tag is cut; no .go code is touched. go build and go vet are both clean. Windows targets were verified locally to produce valid PE32+ executables (amd64 and arm64).
6 tasks
chazmaniandinkle
pushed a commit
that referenced
this pull request
Apr 22, 2026
… gap Closes Agent F gap #3 (session management, CRITICAL) — the last of the eight critical MCP surface gaps. Implements the hybrid design in cog://mem/semantic/surveys/2026-04-21-consolidation/ agent-P-session-management-evaluation with a few user-approved amendments (see below). Kernel changes -------------- - internal/engine/sessions.go: typed SessionState, SessionRegistry, HandoffState, HandoffRegistry with RWMutex / Mutex guards; session-id format validation; idempotent-register "update semantics"; atomic first-wins claim with TTL enforcement; replay-from-bus at startup so the in-memory view rebuilds from bus_sessions + bus_handoffs. - internal/engine/serve_sessions_mgmt.go: 8 HTTP handlers — POST /v1/sessions/register POST /v1/sessions/{id}/heartbeat POST /v1/sessions/{id}/end GET /v1/sessions/presence POST /v1/handoffs/offer GET /v1/handoffs POST /v1/handoffs/{id}/claim POST /v1/handoffs/{id}/complete The existing /v1/sessions and /v1/sessions/{id}[/context] routes (TAA inference context, regression-locked) are preserved untouched; the new specific patterns coexist thanks to Go 1.22 method-aware routing. - internal/engine/mcp_sessions.go: 8 cog_* MCP tools over the same registries so a future native client (Wave widget, desktop app, cog CLI) can use handoff without the Python bridge (amendment #5 — two MCP surfaces coexist by design). - internal/engine/sessions_test.go: 15 unit + integration tests (validation, lifecycle 404/409, active-window presence, task-field validation, 8-way concurrent claim atomicity, TTL expiry, phantom offer, complete-without-claim, replay rebuilds state, claim_rejected observability, end-to-end MCP round-trip). Amendments applied vs the survey -------------------------------- 1. No parallel coexistence. All four consumers are in-tree (this PR, the bridge on a local branch, the skill doc, and cmd_bus.go); migrated atomically. The survey's Open Question #1 was skipped. 2. Idempotent register = update-semantics (survey's Open Question #2 recommendation). Re-register during the active window updates the in-memory row; re-register after end is allowed if the prior row is ended or its heartbeat is outside the active window. 3. `handoff.claim_rejected` event emitted on every rejected claim, with reason ∈ {already_claimed, ttl_expired, offer_not_found, out_of_order}, attempting_session, and conflicting_session when relevant. Cheap; big audit value (amendment #4). 4. Two MCP surfaces coexist by design — 8 cogos_* bridge tools over the Python sandbox + 8 cog_* kernel-native tools via /mcp. Both hit the same kernel registries (amendment #5). Bridge migration ---------------- A paired local branch on cog-sandbox-mcp (`feat/sessions-kernel-native- bridge`, NOT pushed) refactors the 8 cogos_* tools to shim over the new kernel routes, removes client-side aggregation, and rewrites tests/test_session_handoff.py for the new wire shape. Bridge MCP signatures and the never-raise {"success": False, "error": ..., "bus_id"} envelope are preserved — no breaking change for agents using the bridge. Testing ------- - `go build ./...`, `go vet ./...`: silent. - `go test ./internal/engine/... -short -race -count=1`: green (pre-existing + new suite passes under race detector). - End-to-end smoke on port 6932 with a test workspace: register → heartbeat → offer → list → claim → second claim (→ 409 + claim_rejected event) → complete. bus_sessions chain: 3 events. bus_handoffs chain: 4 events (offer, claim, claim_rejected, complete). Bridge tools replayed the same flow against the live kernel with back-compat response shapes intact. - The running kernel at :6931 was NOT touched during this work. Survey reference: cog://mem/semantic/surveys/2026-04-21-consolidation/ agent-P-session-management-evaluation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Complete implementation of the CogOS externalized attention and executive function modulation (EA/EFM) architecture across 10 commits:
Test plan
go build ./...passesgo test ./internal/engine/... -short)TestClaudeCodeBuildPromptIncludesMultipleUserTurnsfailure is unchanged (not caused by these changes)🤖 Generated with Claude Code + Codex