agent-agentic-os

Bootstrap the project by triggering the agentic-os-setup conversational architect.

Force garbage collection and conflict resolution on the tiered memory system.

Trigger with "use the agentic-os-setup agent", "run the setup agent", "set up an agentic OS", "persist memory", "add the OS harness", or when the user requires memory persistence, repository-level conventions, or autonomous background loops. Directs the orchestration, synthesis, and provisioning of a persistent AI environment. <example> Context: User wants to initialize their project for AI agents. user: "Can you help me set up an agentic OS in this folder?" assistant: "I'll use the agentic-os-setup agent to handle the full orchestration for you." <commentary> User requesting specific specialized task execution. Trigger agent. </commentary> </example> <example> Context: A non-technical user wants the AI to remember things. user: "How do I get Claude to persist its memory in my repo between sessions?" assistant: "I'll launch the agentic-os-setup agent to scaffold a persistent memory environment for you." <commentary> User asking for a core Agentic OS feature (persistence). Trigger agent. </commentary> </example> <example> Context: User has an existing codebase but no .claude config. user: "I already have a big project here, can you just add the OS harness without breaking it?" assistant: "Yes, I will run the agentic-os-setup agent to carefully layer the Agentic OS into your existing project." <commentary> Partial setup / integration requested. Trigger agent. </commentary> </example>

Trigger with "run health check", "check os metrics", "system monitor", or when the user wants to review the Agentic OS liveness metrics across the Event Bus, locks, and memory arrays. <example> user: "Run a system monitor check on the OS." assistant: "I'll execute the os-health-check agent to scan the event bus and state file." <commentary> User explicitly requested a system diagnostic, triggering the health check agent. </commentary> </example>

Interactive entry point for starting a skill evaluation loop via the Triple-Loop Learning System. Trigger with "eval [skill]", "evaluate [skill]", "run eval on [skill]", "setup triple-loop lab for [skill]". Handles full setup using the canonical Sibling Repo Labs protocol (creates an isolated repo for safe iteration). <example> Context: User wants to start an eval loop on a skill safely. user: "eval using-git-worktrees" assistant: [triggers triple-loop-architect, resolves skill path, scaffolds sibling lab repo, prepares evals] </example>

Unattended overnight Triple-Loop Learning orchestrator. Oversees the autonomous INNER looping (Strategic Double-Loop and Tactical Single-Loop) on a target skill in its isolated sibling lab. Uses Gemini or Copilot CLI for proposals, gated strictly by objective `evaluate.py` performance. Trigger with "trigger the triple-loop-orchestrator on [skill] for [N] iterations", or "run orchestrator all night on [skill]". <example> Context: User wants to improve a skill headlessly. user: "Trigger triple-loop-orchestrator on link-checker for 80 iterations." assistant: "Launching the Triple-Loop Orchestrator to oversee unattended iterations on the link-checker lab..." </example>

Trigger with "/os-clean-locks", "clear all locks", "reset agent locks", or when an agent is deadlocked and cannot acquire a lock because a previous agent crashed and left a stale lock behind in `context/.locks/`. <example> Context: User is seeing errors about locks already existing. user: "/os-clean-locks" assistant: <Bash> rm -r context/.locks/ python3 context/kernel.py state_update active_agent os-clean-locks </Bash> </example> <example> Context: Agent detects a deadlock when trying to acquire a lock during a task. assistant: [autonomously] "The acquire_lock call for 'memory' failed -- a prior agent likely crashed and left a stale lock. I'll invoke os-clean-locks to clear it before retrying." <commentary> Implicit audit trigger -- agent detects deadlock from kernel output and self-heals using os-clean-locks without user prompting. </commentary> </example>

Reviews a completed os-eval-runner lab run and backports approved changes to master plugin sources. Trigger with "backport the eval results", "review the lab run", "apply eval improvements to master", "check what the eval agent changed".

Bootstraps a skill evaluation lab repo for an autoresearch improvement run. Trigger with "set up an eval lab", "bootstrap the eval repo", "prepare the test repo for skill evaluation", "create an eval environment for this skill", "set up the lab space for this skill", or when starting a new skill optimization run that needs a standalone test environment. <example> Context: User wants to start an improvement run on a skill in an isolated lab repo. user: "Set up an eval lab for the link-checker skill" assistant: [triggers os-eval-lab, runs intake interview, bootstraps lab repo, installs engine, copies plugin files, generates eval-instructions.md] </example> <example> Context: User has a lab repo but needs it configured. user: "Prepare the test repo at <USER_HOME>/Projects/test-my-skill-eval for skill evaluation" assistant: [triggers os-eval-lab, installs engine, copies plugin files, generates eval-instructions.md] </example>

Trigger: "evaluate this skill", "run autoresearch loop on", "optimize this skill". Use when an agent proposes a change to an existing skill and needs empirical validation. <example> Context: Start autonomous improvement loop on a skill. user: "Run the autoresearch loop on <SKILL_PATH> for 20 iterations" assistant: [triggers os-eval-runner, runs Mode 1 intake] </example> <example> Context: Incomplete optimize request. user: "Optimize the commit skill" assistant: [triggers os-eval-runner, runs Phase 0 intake interview] </example> <example> Context: `Triple-Loop Retrospective` proposes a skill edit. assistant: [autonomously] "Before I apply this description change, I'll run os-eval-runner to confirm." </example> <example> Context: An agent is asking for general information about a skill, not evaluating a proposed change. agentic-os-setup: "Tell me about the os-clean-locks skill." assistant: "It cleans up stale lock files..." </example>

Trigger with "explain agentic os", "how do I set up a persistent agent environment", "what is the CLAUDE.md hierarchy", "explain the context folder structure", "how does session memory work", "what is soul.md or user.md", "explain auto-memory or MEMORY.md", "what is a loop scheduler or heartbeat", or when the user asks for the canonical guide.

Pattern 5: Concurrent Event-Driven Multi-Agent Loop. Coordinates multiple Claude sessions as OS threads sharing a common event bus and memory address space. Every loop cycle is a full improvement cycle: execute, eval against benchmark (KEEP/DISCARD), emit friction events during work, close with post_run_metrics, agent self-assessment survey saved to retrospectives, memory persistence, and Triple-Loop Retrospective trigger if friction threshold crossed. Four coordination topologies: turn-signal, fan-out, request-reply, triple-loop (Pattern D).

Trigger with "show me the improvement chart", "how are we improving", "progress report", "graph the eval scores", "show cycle of improvement", "what's the trend", "are we getting better". Produces a visual/text summary of how the agentic loop is improving across cycles. Do NOT use this to run the learning loop or evaluate a specific skill change.

Trigger: "set up agentic OS", "initialize agent harness", "init my project for AI agents", "where do I put CLAUDE.md", "create my agent environment", "set up persistent memory". Guides users through an interview to understand their use case, then scaffolds the right Agentic OS structure. Use even when the user just asks WHERE to put files.

Trigger with "remember this", "update memory", "what should we record from this session", "capture learnings", "write a session log", or when closing a session. Guides agents on managing memory hygiene across sessions, deciding what to write to dated memory logs, what to promote to long-term memory.md, and when to archive. <example> User: I'm done for the day, can you write up a session log? Agent: <Bash> python3 context/kernel.py emit_event --agent os-memory-manager --type intent --action promote_memory python3 context/kernel.py state_update active_agent os-memory-manager </Bash> </example> <example> User: That's all, logging off now. Agent: <Bash> python3 context/kernel.py acquire_lock memory </Bash> </example> <example> User: How does the memory system work? Agent: <Read> ./references/architecture/context-folder-patterns.md </Read> </example>

Continuously improves an existing agent skill based on eval results using the RED-GREEN-REFACTOR cycle. Apply when a skill's routing accuracy is low, trigger descriptions need sharpening, or os-eval-runner scores are below target. (1) run a RED baseline to observe the failure mode, (2) apply a focused patch and verify with os-eval-runner (GREEN), (3) refactor to close loopholes until score meets threshold. Integrates with os-eval-runner as the objective eval gate. NOT for scaffolding new skills — use create-skill (agent-scaffolders) for that.

Audit a file for TODO comments, pending work items, or technical debt markers. Useful for checking code readiness before a commit or reviewing task status. Trigger with "check for todos", "audit for debt", "list pending work", or "scan for TODOs".

3 hooks across 3 events