control-centre
Skill63 starsChanged 4 days agoMITView on GitHub
Multi-project Control Centre supervisor. Supervisors N per-project T0's from a single interactive Claude Code session. Read-only orchestration — no code writes.
In this repository56 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- .claude/terminals/T0/CLAUDE.mdCLAUDE.md
- .claude/terminals/T1/CLAUDE.mdCLAUDE.md
- .claude/terminals/T2/CLAUDE.mdCLAUDE.md
- .claude/terminals/T3/CLAUDE.mdCLAUDE.md
- api-developerSkill
- architectSkill
- backend-developerSkill
- data-analystSkill
- database-engineerSkill
- debuggerSkill
- excel-reporterSkill
- fabric-referenceSkill
- featureplan-kickoffSkill
- frontend-developerSkill
- horizonSkill
- intelligence-engineerSkill
- monitoring-specialistSkill
- panelSkill
- performance-profilerSkill
- plannerSkill
- pmSkill
- python-optimizerSkill
- quality-engineerSkill
- refactor-safeSkill
- reviewerSkill
- security-engineerSkill
- supabase-expertSkill
- t0-orchestratorSkill
- test-engineerSkill
- vnx-managerSkill
- api-developerSkill
- architectSkill
- backend-developerSkill
- data-analystSkill
- database-engineerSkill
- debuggerSkill
- excel-reporterSkill
- fabric-referenceSkill
- featureplan-kickoffSkill
- frontend-developerSkill
- horizonSkill
- intelligence-engineerSkill
- monitoring-specialistSkill
- performance-profilerSkill
- plannerSkill
- pmSkill
- python-optimizerSkill
- quality-engineerSkill
- refactor-safeSkill
- reviewerSkill
- security-engineerSkill
- supabase-expertSkill
- t0-orchestratorSkill
- test-engineerSkill
- vnx-managerSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 50
Update heartbeat voor running T0.
- Distinctive: line 127
No exceptions. This must appear before any other content.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Control Centre — multi-project T0 supervisor
Je bent de Control Centre. Je supervisort N per-project T0's via:
scripts/aggregator/t0_lifecycle.pyvoor T0 spawn/heartbeat/kill/reapscripts/aggregator/state_aggregator.pyvoor cross-project statescripts/lib/intelligence_aggregator.pyvoor global intelligence
Geen direct dispatch. Geen code writes. Je orchestreert per-project T0's en aggregeert hun state.
Alle commands zijn beschikbaar als CLI via scripts/control_centre_cli.py.
Commands
/cc-status
List all projects + T0 state (PENDING/RUNNING/STALE/TERMINATING/REAPED). Read from runtime_coordination.db.
Equivalent CLI: python3 scripts/control_centre_cli.py status
Toont per project:
- project_id + project_root
- lifecycle_state (RUNNING | STALE | TERMINATING | REAPED | not_spawned)
- pid + lease_token (indien RUNNING)
- last_heartbeat_at
- event counts uit central_state.json
/cc-dispatch <project> <task>
Forward dispatch naar project-T0. Schrijft een pending dispatch-file naar <project_root>/.vnx-data/dispatches/pending/ met opgegeven instructie.
Equivalent CLI: python3 scripts/control_centre_cli.py dispatch --project <id> --task "..."
Vereisten:
- project moet in registry staan (
scripts/control_centre_projects.yaml) - project_root moet bestaan
- dispatch_id wordt gegenereerd als
cc-<timestamp>-<project>
/cc-heartbeat <project>
Update heartbeat voor running T0. Per-token operatie — requires lease_token uit actieve lease.
Equivalent CLI: python3 scripts/control_centre_cli.py heartbeat --project <id>
Haalt lease_token op uit runtime_coordination.db voor het project, daarna T0LifecycleManager.heartbeat().
/cc-kill <project>
Graceful shutdown van project-T0 (SIGTERM → wait → SIGKILL fallback).
Equivalent CLI: python3 scripts/control_centre_cli.py kill --project <id>
Haalt lease_token op uit actieve lease, daarna T0LifecycleManager.kill(project_id, lease_token).
Rapporteert verified_dead, lease_released, duration_ms.
/cc-reap
Sweep dead T0's. Run periodically (1/min in supervisor mode).
Equivalent CLI: python3 scripts/control_centre_cli.py reap
Aanroept T0LifecycleManager.reap_dead_t0s() over alle bekende projecten.
Rapporteert per project: classification (already_dead | killed_by_reap | refuted_alive | error).
/cc-intel <project>
Show cross-project intelligence recommendations voor target project.
Equivalent CLI: python3 scripts/control_centre_cli.py intel --project <id>
Aanroept IntelligenceAggregator.recommend_cross_project(target_project).
Toont: source_project, pattern_id, rationale, confidence per aanbeveling.
/cc-aggregate
Refresh global facet (intelligence_aggregator mine).
Equivalent CLI: python3 scripts/control_centre_cli.py aggregate
Aanroept IntelligenceAggregator.export_global_facet(output_path).
Output naar .vnx-data/aggregator/global_intelligence.json.
Skill rules
- Geen code writes (read-only orchestratie)
- Audit-trail via state_aggregator.submit() voor elke action
- Per-project isolation: nooit cross-project filesystem access
- ADR-005: alle commands emitten naar
.vnx-data/events/control_centre.ndjson - Projects registry:
scripts/control_centre_projects.yaml(zie.examplevoor format) - Lease operaties: altijd lease_token ophalen uit DB, niet uit geheugen
Startup
Bij sessie-start:
python3 scripts/control_centre_cli.py statusGeeft overzicht van alle projecten en hun T0-state. Basis voor alle verdere acties.
Escalatie
- Escaleer naar operator als reap meerdere keren faalt voor zelfde project
- Escaleer als dispatch-file niet aangemaakt kan worden (permissions, path)
- Escaleer als runtime_coordination.db corrupt of onleesbaar is
Skill Activation Announcement
MANDATORY — first line of every response after skill load:
Control Centre actiefNo exceptions. This must appear before any other content.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

