{"id":"io.github.eidetic-works/nucleus","slug":"io-github-eidetic-works-nucleus","title":"Nucleus","description":"Persistent memory for AI agents — shared .brain across Claude Code, Cursor & Codex","version":"1.17.1","publisher":{"kind":"github","id":"eidetic-works","label":"@eidetic-works on GitHub","url":"https://github.com/eidetic-works"},"website":"https://nucleusos.dev","repository":"https://github.com/eidetic-works/nucleus-mcp","remotes":[{"type":"streamable-http","url":"https://relay.nucleusos.dev/mcp"}],"packages":[{"registryType":"pypi","identifier":"nucleus-mcp","version":"1.17.1","runtimeHint":"uvx","transport":{"type":"stdio"},"environmentVariables":[{"description":"Path to .brain/ directory (defaults to .brain in current directory)","format":"string","name":"NUCLEUS_BRAIN_PATH"}]}],"published":"2026-11-02","updated":"2026-11-02","status":"online","auth":"","latency_ms":2274,"checked":"2026-11-19","uptime":100,"checks":3,"tools":[{"name":"nucleus_governance","title":"Governance","description":"Governance, Hypervisor & security tools for the Nucleus Agent OS.\n\nActions:\n  auto_fix_loop   - Auto-fix loop: Verify->Diagnose->Fix->Verify (3 retries). params: {file_path, verification_command}\n  lock            - [HYPERVISOR] Lock a file/dir immutable (chflags uchg). params: {path}\n  unlock          - [HYPERVISOR] Unlock a file/dir. params: {path}\n  set_mode        - [HYPERVISOR] Switch IDE con","read_only":false},{"name":"nucleus_features","title":"Feature Flags","description":"Feature tracking, proof generation & MCP server mounting.\n\nActions:\n  add             - Add a feature. params: {product, name, description, source, version, how_to_test, expected_result, status?, tags?}\n  list            - List features. params: {product?, status?, tag?}\n  get             - Get feature by ID. params: {feature_id}\n  update          - Update feature fields. params: {feature_id, stat","read_only":false},{"name":"nucleus_sessions","title":"Session Management","description":"Session management, events, state & checkpoint tools.\n\nActions:\n  save             - Save session for later. params: {context, active_task?, pending_decisions?, breadcrumbs?, next_steps?}\n  resume           - Resume a saved session. params: {session_id?}\n  list             - List all saved sessions\n  check_recent     - Check for recent session to resume (alias: \"current\")\n  end              - End ","read_only":false},{"name":"nucleus_tasks","title":"Task Management","description":"Task management, depth tracking & ADHD context-switch tools.\n\nActions:\n  list              - List tasks. params: {status?, priority?, skill?, claimed_by?, required_role?}\n  get_next          - Get highest-priority unblocked task. params: {skills, required_role?}\n  claim             - Atomically claim a task. params: {task_id, agent_id}\n  update            - Update task fields. params: {task_id, up","read_only":false},{"name":"nucleus_sync","title":"Cross-Agent Sync","description":"Sync, artifact, trigger & deploy management for multi-agent coordination.\n\nActions:\n  identify_agent   - Register agent identity. params: {role, provider, session_id} (per ADR-0005 §D1)\n                     OR legacy {agent_id, environment, role?} (coerced per §D5 until end of Cycle C+2)\n  sync_status      - Check current multi-agent sync status\n  sync_now         - Manually trigger sync. params: ","read_only":false},{"name":"nucleus_next_message","title":"Next Message","description":"Get the next unread relay message. One call — returns full message\nbody and auto-acks. Long-poll (60s timeout).\n\nThis is the agent's idle loop. Call it, get a message or None, act or\nre-call. No separate inbox + ack dance.\n\nThe agent's loop:\n  1. nucleus_next_message() → blocks up to 60s for a message\n  2. If message: handle it (execute task, respond to comment, etc.)\n  3. Re-call nucleus_next_mes","read_only":false},{"name":"nucleus_relay_subscribe","title":"Relay Subscription","description":"Long-poll subscription that pushes ctx.info() on each new inbox file.\n\nReplaces bash polling daemons (watch-relay-*.sh) with server-initiated push.\nCall once at session start (e.g. via SessionStart hook). Server holds the\nsubscription, watches the calling agent's role-specific inbox dir, and\nfires info-level notifications on each new relay file arrival. Client\nre-calls this in a loop for persisten","read_only":true},{"name":"nucleus_ccr_arm","title":"CCR Arm","description":"One-shot convenience: resolve canonical inbox + arm long-poll subscription.\n\nPer PR #2 (CCR server-side auto-arm) — IDE-agnostic relay-arrival arming.\n\nThis is the RECOMMENDED entry point for SessionStart auto-arming across all\nMCP clients (Claude Code, Antigravity, Cursor, Windsurf, etc.). Equivalent to:\n    1. resolve_canonical_inbox_name(role) → canonical inbox name\n    2. nucleus_relay_subscri","read_only":false},{"name":"nucleus_wakeup_wait","title":"Task Wakeup Wait","description":"Quick scan for a PENDING task. Returns the task directly if one is\navailable, or None if no task is ready within the timeout.\n\nDefault timeout is 5s (non-blocking). The agent should NOT loop on\nthis — tasks arrive via relay push. This is a fallback for when the\nagent wants to check for tasks without waiting for a relay.\n\nNo args needed — the role is auto-detected from posture\n(.brain/posture/curre","read_only":false},{"name":"nucleus_orchestration","title":"Orchestration","description":"Satellite view, commitments, loops, patterns & metrics.\n\nActions:\n  satellite         - Unified satellite view. params: {detail_level?}\n  scan_commitments  - Scan artifacts for new commitments\n  archive_stale     - Auto-archive commitments older than 30 days\n  export            - Export brain to zip\n  list_commitments  - List open commitments. params: {tier?}\n  close_commitment  - Close a commitme","read_only":false},{"name":"nucleus_telemetry","title":"Telemetry","description":"LLM tiers, telemetry, PEFS notifications & protocol tools.\n\nActions:\n  set_llm_tier         - Set default LLM tier. params: {tier}\n  get_llm_status       - Get LLM tier configuration\n  record_interaction   - Record user interaction timestamp\n  value_ratio          - Get Value Ratio metric\n  check_kill_switch    - Check Kill Switch status\n  pause_notifications  - Pause PEFS notifications\n  resume_n","read_only":false},{"name":"nucleus_slots","title":"Slots & Sprints","description":"Orchestration slots, sprints & mission management.\n\nActions:\n  orchestrate        - THE GOD COMMAND. params: {slot_id?, model?, alias?, mode?}\n  slot_complete      - Mark task complete. params: {slot_id, task_id, outcome?, notes?}\n  slot_exhaust       - Mark slot exhausted. params: {slot_id, reset_hours?}\n  status_dashboard   - ASCII dashboard. params: {detail_level?}\n  autopilot_sprint   - Sprint","read_only":false},{"name":"nucleus_infra","title":"Infrastructure","description":"Infrastructure: file changes, cloud, marketing & strategy tools.\n\nActions:\n  file_changes        - Get pending file change events\n  gcloud_status       - Check GCloud auth status\n  gcloud_services     - List Cloud Run services. params: {project?, region?}\n  list_services       - List Render.com services\n  scan_marketing_log  - Scan marketing log for failures\n  synthesize_strategy - Analyze marketi","read_only":false},{"name":"nucleus_agents","title":"Agent Spawning","description":"Agent spawning, critic, swarm, memory, ingestion & dashboard tools.\n\nActions:\n  spawn_agent          - Spawn Ephemeral Agent. params: {intent, execute_now?, persona?, confirm?}. HITL: requires confirm=true.\n  apply_critique       - Apply critique fixes. params: {review_path}\n  orchestrate_swarm    - Start multi-agent swarm. params: {mission, agents?}\n  search_memory        - Search long-term memor","read_only":false},{"name":"nucleus_federation","title":"Federation","description":"Federation management for multi-brain coordination.\n\nActions:\n  status  - Get comprehensive federation status\n  join    - Join a federation via seed peer. params: {seed_peer}\n  leave   - Leave the federation gracefully\n  peers   - List all federation peers with details\n  sync    - Force immediate synchronization with all peers\n  route   - Route a task to the optimal brain. params: {task_id, profil","read_only":false},{"name":"nucleus_engrams","title":"Memory & Engrams","description":"Engrams, health, observability, DSoR & tier system tools.\n\nActions:\n  health              - Get system health status\n  version             - Get Nucleus version info\n  export_schema       - Export MCP toolset as JSON Schema\n  performance_metrics - Get perf metrics. params: {export_to_file?}\n  prometheus_metrics  - Get Prometheus metrics. params: {format?}\n  audit_log           - View cryptographic","read_only":false},{"name":"nucleus_relay","title":"Relay Messaging","description":"Relay-substrate facade — post / read / ack / status.\n\nActions:\n  post    - Send a relay envelope. params: {to, subject, body, sender?,\n            priority?, in_reply_to?, context?, id?, from_session_id?}\n            sender? auto-fills from your session role when omitted.\n            to accepts short role aliases: main, peer, tb, ops, agy, board.\n            \"recipient\" is accepted as an alias for","read_only":false},{"name":"nucleus_route","title":"Cost Router","description":"W5 tier routing — route a prompt to the optimal model tier.\n\nActions:\n  route  - Route a prompt to cheapest capable model.\n           params: {prompt, complexity?, context?, estimated_output_tokens?}\n           complexity ∈ 'routine' | 'complex' | 'sovereign' (default 'routine')\n           Returns: provider, model, cost estimates, sovereignty tier.","read_only":true},{"name":"nucleus_audit","title":"Audit Log","description":"W8 Team-tier tamper-evident audit log (SHA-256 hash chain).\n\nActions:\n  log_event   - Append an audit event to the chain.\n                params: {event_type, actor, resource, outcome, metadata?, team_id?, ts?}\n  query       - Single-tenant query. Rejects team_id='*'.\n                params: {team_id, since?, until?, actor?, event_type?, limit?, offset?}\n  admin_query - Cross-tenant query (team_id","read_only":true},{"name":"nucleus_delegate","title":"Cross-Vendor Delegate","description":"Hand a coding or review task to a cross-vendor lane — a cheap, non-Claude\nagent that does the work for you. Use this INSTEAD OF shelling out to any CLI: it\npicks the vendor, injects the correct per-vendor permission flags, captures all\noutput synchronously, and reports a real status. Do NOT run `agy`/`devin`\nyourself in Bash — the raw CLIs take DIFFERENT per-vendor flags (pass the wrong\none and a ","read_only":false},{"name":"nucleus_lane_init","title":"","description":"Set up an autonomous task-execution loop in a git repo. WHEN TO USE: you have a list of tasks (bugs, features, tests) you want executed automatically without manual prompting — the lane runs watcher/executor/secretary daemons that claim tasks, invoke an LLM to implement them, and independently verify the results. Use this when you want to batch-execute a backlog of well-defined tasks. Creates .bra","read_only":false},{"name":"nucleus_lane_start","title":"","description":"Start autonomous lane daemons (watcher + executor + secretary) in background. WHEN TO USE: after nucleus_lane_init and SPEC.md is written — this launches the loop that autonomously claims and executes tasks. The executor invokes an LLM CLI (devin/agy) per task; the secretary independently verifies each result. Runs until nucleus_lane_stop is called.","read_only":false},{"name":"nucleus_lane_stop","title":"","description":"Stop all running lane daemons. WHEN TO USE: when all tasks are CONFIRMED or you want to pause autonomous execution.","read_only":false},{"name":"nucleus_lane_status","title":"","description":"Check lane status: task counts (PENDING/IN_PROGRESS/DONE/CONFIRMED), daemon health, spec verification. WHEN TO USE: after starting a lane to monitor progress, or before starting to check if a lane is already running.","read_only":false},{"name":"nucleus_lane_feedback","title":"","description":"Report friction/bugs/enhancements to the nucleus team. WHEN TO USE: when the lane misbehaves, a task fails unexpectedly, the SPEC.md format is confusing, or you have an improvement idea. Creates a GitHub issue on the public nucleus-mcp repo (cross-machine visible) AND stores locally. Requires gh CLI auth'd.","read_only":false},{"name":"nucleus_plan_execute","title":"","description":"Execute a plan file autonomously. WHEN TO USE: you have a plan file (.brain/plans/*.md) you want executed autonomously — parses the plan, creates tasks, starts a mission with budget/time limits. Chains import_plan_as_tasks() then _brain_start_mission_impl() and returns the mission status string.","read_only":false},{"name":"nucleus_plan_import","title":"","description":"Import a plan file as PENDING tasks without starting a mission. WHEN TO USE: you have a plan file (.brain/plans/*.md) you want loaded into the task store so the executor daemon / sprint mission can pick the tasks up later — but you don't want to start a mission right now.","read_only":false},{"name":"nucleus_plan_list","title":"","description":"List available plan files in .brain/plans with task counts. WHEN TO USE: you want to discover which plan files exist and how many tasks each contains before importing or executing one.\n\nReturns:\n    JSON string: a list of dicts each with\n    ``{\"name\", \"path\", \"size_bytes\", \"task_count\", \"format\"}``.\n    Missing directory returns ``[]``.","read_only":false}],"server_info":{"name":"Nucleus Brain","version":"3.4.4"},"category":"developer","tags":["memory","cloud","communication"],"kind":"remote","source":"registry","submitted":{},"first_party":false,"domain_rank":null,"score":100,"indexable":true,"first_seen":"2026-02-10","stars":0,"badges":[],"sponsor":null,"page":"https://agentsindex.org/agents/io-github-eidetic-works-nucleus/"}