Claude Code Skills · page 87

Use when diagnosing runaway processes with the pt wrapper and choosing safe remediation.

Use when offloading slow builds to remote workers or recovering RCH worker, hook, SSH, sync, or disk issues.

Use when applying a skill repeatedly with progressive deepening for iterative improvement.

>-

Use when using ru for multi-repo commits, sync, GitHub review, or maintenance automation.

>-

Use when configuring SSH access, keys, tunnels, host diagnostics, or safe remote command workflows.

Use when restoring machine responsiveness from high CPU, memory, IO, cache, or runaway process pressure.

Use when reviewing code with UBS for bugs, security issues, AI-generated quality, or pre-commit checks.

Use when tending NTM agent swarms, unsticking panes, handling rate limits, or coordinating convergence.

Explain AgentOps workflows.

Run AGY headlessly via scheduled ticks or `agy -p`, capture agentapi JSONL evidence, and validate automated AGY loops or event streams.

|-

Drive AgentOps in AGY: loop, plugins, memory, evidence, scoped worktrees.

|-

Manage the PROGRAM.md/AUTODEV.md contract consumed by evolve/factory ticks. Use for loop rules, boundaries, or PROGRAM.md repair.

Track br issues and dependencies for Codex agents. Triggers: "beads", "br", "track issue", "create task", "find work", "ready issues".

Initialize AgentOps project files.

Separate goals from implementation.

Compile .agents knowledge wiki.

Multi-perspective review for Codex using the current sub-agent runtime. Triggers: "council", "get consensus", "multi-model review", "multi-perspective review", "council validate", "council brainstorm", "council research".

Hands-free epic execution for Codex using wave-based sub-agents and lead-side validation. Triggers: "crank", "run epic", "execute epic", "run all tasks", "hands-free execution", "crank it".

Validate product fit before discovery. Use when: framing a problem, checking product/market fit, or pressure-testing user value before writing a discovery packet or any code.

Create dense execution packets.

Run autonomous improvement loops.

Check knowledge flywheel health.

Mine transcripts into learnings.

Maintain AgentOps goals.

Write compact session handoffs.

Implement one tracked issue.

Load relevant .agents context.

Install or run the seven-move operating-loop Workflow for AgentOps plugin users and multi-agent orchestration.

Decompose goals into issue plans.

Review completed work and learn. Use when: a task, PR arc, or session is finished and you want to extract learnings, or after ≥5 PRs (the scope checkpoint).

Stress-test plans before work. Use when: a plan is drafted but not yet executed and you want to surface failure modes, risks, and what would prove it wrong before committing.

Validate, commit, and push.

Record Brownian Ratchet gates.

Recover session context.

Probe docs and skills. Use when: adversarially probing a doc, skill, plan, or claim for weaknesses, gaps, or unstated assumptions before it ships.

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

>

Deep codebase exploration with multi-angle analysis for Codex agents. Output: .agents/research/*.md. Triggers: "research", "investigate", "explore codebase", "how does X work".

Run discovery, crank, validation.

Hard-block edits outside declared frozen directories and protect paths during risky changes.

Universal AgentOps init prompt for starting or onboarding a fresh agent session.

Audit SKILL.md files against the AgentOps template and readiness checks. Use for quality reviews or template compliance.

Scaffold or absorb new SKILL.md files against the unified AgentOps template. Triggers: "create a skill", "scaffold skill", "absorb external skill", "new skill".

Show AgentOps work status.

Spawn isolated Codex sub-agents for parallel task execution using the current runtime primitives. Triggers: "swarm", "spawn agents", "parallel work", "run in parallel", "parallel execution".

Produce PASS/WARN/FAIL verdicts for artifacts, plans, code, PRs, or gates. Use when: you need a structured verdict on an artifact, plan, code, PR, or CI gate before proceeding.

Switch coding-agent accounts on a usage/rate limit or to spread swarm lanes. Routes by host+agent: macOS+Claude via claude-acct; Codex/Gemini and Linux/WSL via caam. Triggers: "account-rotation", "account rotation", "switch coding-agent accounts on a".

Use when operating ACFS flywheel health checks, init, and agent loop tooling from ~/acfs/bin/acfs.

Make an out-of-session agent AgentOps-native with skills, the ao CLI, local cockpit proof, and CI backstop telemetry instead of runtime hooks. Triggers: "agent-native", "agent native", "make an out-of-session agent agentops-native".

Analyzes, generates, and enhances CLAUDE.md files for any project type using best practices, modular architecture support, and tech stack customization. Use when setting up new projects, improving existing CLAUDE.md files, or establishing AI-assisted development standards.

Re-detect this project's tech stack from package.json / requirements.txt / pyproject.toml / go.mod / Cargo.toml and diff it against the Tech Stack section of every CLAUDE.md. Read-only — returns added / removed / renamed dependencies, never edits.

Audit every CLAUDE.md in this project for drift against the last week of git history. Flags sections that reference deleted files, renamed paths, or removed dependencies. Read-only — returns a punch list, never edits.