Skip to content

Skills

Skills are the composable units of AgentOps. Each one is a declarative capability — a prompt contract with optional scripts, references, and enforced metadata — that any compatible harness (Claude Code, Codex, OpenCode) can invoke.

69 skills ship with AgentOps. Start with the headline nine below, then explore the full catalog when you need more specialized tools.

Headline skills — use these first

Skill Use it when
quickstart You want the fastest setup check and next action
council You want independent judges to review a plan, PR, or decision
research You need codebase context and prior learnings before changing code
pre-mortem You want to pressure-test a plan before implementation
implement You want one scoped task built and validated
rpi You want discovery, build, validation, and bookkeeping in one flow
vibe You want a code-quality and risk review before shipping
evolve You want a goal-driven improvement loop with regression gates
dream You want overnight knowledge compounding that never mutates source code

Which skill do I need next?

See the Decision Tree for a visual walkthrough, or SKILL-ROUTER for rule-based routing.

Complete catalog by family

Every skill the system ships with, grouped by purpose:

Validation

  • council — Multi-model consensus council. Spawns parallel judges with configurable perspectives. Modes: validate, brainstorm, research. Triggers: "council", "get consensus", "multi-model review", "multi-perspective review", "council validate", "council brainstorm", "council research".
  • vibe — Comprehensive code validation. Runs complexity analysis then multi-model council. Answer: Is this code ready to ship? Triggers: "vibe", "validate code", "check code", "review code", "code quality", "is this ready".
  • pre-mortem — Validate a plan or spec before implementation using multi-model council. Answer: Is this good enough to implement? Triggers: "pre-mortem", "validate plan", "validate spec", "is this ready".
  • post-mortem — Wrap up completed work. Council validates the implementation, then extract and process learnings. Triggers: "post-mortem", "wrap up", "close epic", "what did we learn".
  • red-team — >-

Flows

  • research — Deep codebase exploration. Triggers: research, explore, investigate, understand, deep dive, current state.
  • plan — Epic decomposition into trackable issues. Triggers: "create a plan", "plan implementation", "break down into tasks", "decompose into features", "create beads issues from research", "what issues should we create", "plan out the work".
  • implement — Execute a single issue with full lifecycle. Triggers: "implement", "work on task", "build this", "start feature", "pick up next issue", "work on issue".
  • crank — Hands-free epic execution. Runs until ALL children are CLOSED. Uses /swarm with runtime-native spawning (Codex sub-agents or Claude teams). NO human prompts, NO stopping. Triggers: "crank", "run epic", "execute epic", "run all tasks", "hands-free execution", "crank it".
  • swarm — Spawn isolated agents for parallel task execution. Auto-selects runtime-native teams (Claude Native Teams in Claude sessions, Codex sub-agents in Codex sessions). Triggers: "swarm", "spawn agents", "parallel work", "run in parallel", "parallel execution".
  • rpi — Full RPI lifecycle orchestrator. Delegates to /discovery, /crank, /validation phase skills. One command, full lifecycle with complexity classification, --from routing, and optional loop. Triggers: "rpi", "full lifecycle", "research plan implement", "end to end".
  • evolve — Goal-driven v2 autonomous improvement loop. Runs the post-mortem, repo analysis, next-work selection, plan/pre-mortem, implementation, validation, and repeat cadence through /rpi and ao evolve. Also pulls from open beads when goals all pass. Use when you want to "improve", "iterate", "fix issues", "work through tasks", "evolve", "check goal fitness", "run improvement loop", "pick up next work", or "postmortem and continue".
  • discovery — Full discovery phase orchestrator. Brainstorm + ao search + research + plan + pre-mortem gate. Produces epic-id and execution-packet for /crank. Triggers: "discovery", "discover", "explore and plan", "research and plan", "discovery phase".
  • validation — Full validation phase orchestrator. Vibe + post-mortem + retro + forge. Reviews implementation quality, extracts learnings, feeds the knowledge flywheel. Triggers: "validation", "validate", "validate work", "review and learn", "validation phase", "post-implementation review".

Bookkeeping

  • retro — Quick-capture a learning. For full retrospectives, use /post-mortem. Trigger phrases: "quick learning", "capture lesson", "retro quick".
  • forge — Mine transcripts for knowledge - decisions, learnings, failures, patterns. Triggers: "forge insights", "mine transcripts", "extract knowledge".
  • flywheel — Knowledge flywheel health monitoring. Checks velocity, pool depths, staleness. Triggers: "flywheel status", "knowledge health", "is knowledge compounding".
  • compile — >
  • harvest — >
  • inject — Inject relevant knowledge into session context from .agents/ artifacts. Triggers: "inject knowledge", "recall context", SessionStart hook.
  • provenance — Trace knowledge artifact lineage and sources. Find orphans, stale citations. Triggers: "where did this come from", "trace this learning", "knowledge lineage".

Session

  • handoff — Create structured handoff for session continuation. Triggers: handoff, pause, save context, end session, pick up later, continue later.
  • recover — Post-compaction context recovery. Detects in-progress RPI and evolve sessions, loads knowledge, shows recent work and pending tasks. In Codex v0.115.0+, native hooks handle lifecycle automatically; for older versions, it prefers the explicit hookless fallback path. Triggers: "recover", "lost context", "where was I", "what was I working on".
  • status — Single-screen dashboard showing current work, recent validations, flywheel health, and suggested next action. Triggers: "status", "dashboard", "what am I working on", "where was I".
  • trace — Trace design decisions and concepts through session history, handoffs, and git. Triggers: "trace decision", "how did we decide", "where did this come from", "design provenance", "decision history".
  • dream — >
  • using-agentops — Meta skill explaining the AgentOps operating model. Hook-capable runtimes inject it at session start; Codex uses it through the explicit startup fallback. Covers bookkeeping, validation, primitives, flows, the RPI lifecycle, and the skill catalog.

Product

  • product — Interactive PRODUCT.md generation. Interviews you about mission, personas, value props, and competitive landscape, then generates a filled-in PRODUCT.md. Triggers: "product", "create product doc", "product definition", "who is this for".
  • goals — Maintain GOALS.yaml and GOALS.md fitness specifications. Measure fitness, manage directives, track drift, add/prune goals. Triggers: "goals", "goal status", "show goals", "add goals", "prune goals", "clean goals", "goal drift", "goal history", "export goals", "meta goals", "migrate goals".
  • release — Release your software. Pre-flight validation, changelog generation, version bumps, release commit, tag, curated release notes. Boundary: everything up to the git tag. Triggers: "release", "cut a release", "prepare release", "release check".
  • readme — Generate a gold-standard README for any project. Interviews you about the problem, generates a draft following battle-tested patterns, then council-validates it. Triggers: "readme", "write readme", "generate readme", "improve readme", "rewrite readme".
  • doc — Generates, validates, and syncs documentation for any repository type. Produces code-maps, checks doc coverage, finds missing docs, and validates existing documentation against code. Triggers: doc, documentation, code-map, doc coverage, validate docs, generate docs, sync docs, update docs, find missing docs.
  • oss-docs — Scaffold and audit OSS documentation packs for open source projects. Triggers: "add OSS docs", "setup contributing guide", "add changelog", "prepare for open source", "add AGENTS.md", "OSS documentation".

Utility

  • brainstorm — Separate WHAT from HOW before planning. Clarify goals, explore approaches, capture structured design decisions. Triggers: brainstorm, explore idea, clarify goal, idea phase.
  • bug-hunt — Investigate suspected bugs or run proactive code audits. Triggers: "bug", "broken", "doesn''t work", "failing", "investigate bug", "debug", "find the bug", "troubleshoot", "audit code", "find bugs in", "code audit", "hunt bugs".
  • complexity — Analyze code complexity and find refactor targets using radon/gocyclo. Triggers: "complexity", "analyze complexity", "find complex code", "refactor targets", "cyclomatic complexity", "code metrics".
  • scaffold — Project scaffolding, component generation, and boilerplate setup. Triggers: "scaffold", "new project", "init project", "create project", "generate component", "setup project", "starter", "boilerplate".
  • push — Test, commit, and push in one atomic workflow. Runs Go and Python tests, commits with conventional message, pushes to current branch.
  • refactor — Safe, verified refactoring with regression testing at each step. Identify targets, plan transformation, execute incrementally. Triggers: "refactor", "restructure", "extract", "rename", "move", "simplify", "reduce complexity", "clean up", "decompose".
  • test — Test generation, coverage analysis, and TDD workflow. Triggers: "test", "generate tests", "test coverage", "write tests", "tdd", "add tests", "test strategy", "missing tests", "coverage gaps".
  • deps — Dependency audit, update, vulnerability scanning, and license compliance. Triggers: "deps", "dependencies", "update deps", "audit dependencies", "dependency update", "vulnerable dependencies", "outdated packages", "license check", "dep audit".
  • perf — Performance profiling, benchmarking, regression detection, and optimization. Triggers: "perf", "performance", "benchmark", "profile", "slow", "optimize", "latency", "throughput", "memory leak", "perf regression".
  • review — Review incoming PRs, agent-generated changes, or diffs. Structured review with security, correctness, performance, and maintainability checks. Triggers: "review", "review PR", "review changes", "code review", "review this PR", "review agent output", "check this diff".
  • security — Continuous repository security scanning and release gating. Triggers: "security scan", "security audit", "pre-release security", "run scanners", "check vulnerabilities".
  • security-suite — Composable security suite for binary and prompt-surface assurance, static analysis, dynamic tracing, repo-native redteam scans, contract capture, baseline drift, and policy gating. Triggers: "binary security", "reverse engineer binary", "black-box binary test", "behavioral trace", "baseline diff", "prompt redteam", "security suite".

Platform

  • beads — Manages git-based issue tracking using the bd CLI: creates issues, tracks blockers, routes work across rigs (independent workstreams with their own issue prefixes), and organizes beads (issues) hierarchically with parent-child dependencies. Beads marked "slingable" are ready to hand off between agents or sessions. Use when: "track issues", "create beads issue", "show blockers", "what''s ready to work on", "beads routing", "prefix routing", "cross-rig beads", "slingable beads", or git-based issue tracking with bd.
  • ratchet — Brownian Ratchet progress gates for RPI workflow. Check, record, verify. Triggers: "check gate", "verify progress", "ratchet status".
  • heal-skill — Automated skill maintenance. Detects and fixes common skill issues: missing frontmatter, name mismatches, unlinked references, empty directories, dead references, and Codex parity drift triage. Triggers: "heal-skill", "heal skill", "fix skills", "skill maintenance", "repair skills".
  • update — Reinstall all AgentOps skills globally from the latest source. Triggers: "update skills", "reinstall skills", "sync skills".
  • converter — Cross-platform skill converter. Parse AgentOps skills into a universal bundle format, then convert to target platforms (Codex, Cursor). Triggers: convert, converter, convert skill, export skill, cross-platform.
  • codex-team — Use when you have 2+ tasks that Codex agents should execute. Runtime-native: Codex sub-agents when available, Codex CLI fallback otherwise. Handles file conflicts via merge/wave strategies. Triggers: "codex team", "spawn codex", "codex agents", "use codex for", "codex fix".
  • scenario — Author and manage holdout scenarios for behavioral validation. Scenarios are stored outside the codebase in .agents/holdout/ where implementing agents cannot see them. Triggers: scenario, holdout, behavioral scenario, create scenario, list scenarios.
  • bootstrap — One command to full AgentOps product layer. Sets up GOALS.md, PRODUCT.md, README.md, .agents/ structure, and hooks. Progressive — fills gaps only. Triggers: "bootstrap", "setup agentops", "initialize repo", "full setup".
  • autodev — Manage the PROGRAM.md operational contract that bounds autonomous development loops. Use when setting up or validating PROGRAM.md/AUTODEV.md, choosing between autodev and evolve, or turning a repeated postmortem -> analyze -> plan -> pre-mortem -> implement -> validate cycle into the v2 autonomous workflow.

PR workflow

  • pr-research — Upstream codebase exploration for open source contribution. Outputs contribution guidelines, PR patterns, and maintainer expectations. Triggers: "pr research", "upstream research", "contribution research", "explore upstream repo".
  • pr-plan — Plan an open source PR contribution. Takes pr-research output and produces scope, acceptance criteria, and risk assessment. Triggers: "pr plan", "contribution plan", "plan PR", "plan contribution".
  • pr-implement — Fork-based PR implementation with isolation check. Runs isolation check before starting work. Triggers: "implement PR", "implement contribution", "fork implementation", "code the PR".
  • pr-validate — PR-specific validation: isolation, upstream alignment, quality, scope creep. Triggers: "validate PR", "pr validation", "check PR scope", "scope creep check", "isolation check".
  • pr-prep — PR preparation: git archaeology, test validation, structured PR body generation. Mandatory user review gate before submission. Triggers: "prepare PR", "PR prep", "submit PR", "create PR body", "write PR description".
  • pr-retro — Learn from PR outcomes. Analyzes accept/reject patterns and updates contribution lessons. Triggers: "pr retro", "learn from PR", "PR outcome", "why was PR rejected", "analyze PR feedback".

Other

  • design — Product validation gate for RPI pipeline. Validates goal alignment with PRODUCT.md before discovery. Checks: gap alignment, persona fit, competitive differentiation, precedent, scope boundaries. Council-gated with --preset=product. Triggers: "design", "product validation", "validate product fit", "design gate".
  • grafana-platform-dashboard — Design, refactor, and validate Grafana dashboards for OpenShift/Kubernetes platform operations. Use when users ask to improve platform health dashboards, prioritize critical tenant-impacting signals, filter noise (for example ArgoCD), add Crossplane/Keycloak health panels, validate PromQL programmatically, or apply GrafanaDashboard CR changes live then promote to GitOps.
  • knowledge-activation — Operationalize a mature .agents corpus into usable information. Consolidates packet layers, promotes a belief book, generates playbook candidates, compiles runtime briefings, and surfaces flywheel gaps. Triggers: "operationalize .agents", "turn dot agents into usable information", "knowledge activation", "knowledge flywheel outer loop", "activate knowledge corpus".
  • llm-wiki — Build and maintain a compounding external-knowledge wiki from clipped articles, papers, and transcripts. Triggers: "llm wiki", "ingest this", "second brain", "compile my reading", "wiki lint", "what do we know about ". Based on Andrej Karpathy''s LLM Wiki pattern (April 2026).
  • openai-docs — Use when the user asks how to build with OpenAI products or APIs and needs up-to-date official documentation with citations (for example: Codex, Responses API, Chat Completions, Apps SDK, Agents SDK, Realtime, model capabilities or limits); prioritize OpenAI docs MCP tools and restrict any fallback browsing to official OpenAI domains.
  • quickstart — New user onboarding. Detect setup, explain what AgentOps does, give one next action. Under 30 seconds. Triggers: "quickstart", "get started", "onboarding", "how do I start".
  • reverse-engineer-rpi — Reverse-engineer a product into a feature catalog, code map, and specs. Uses RPI-style loop with verification gates. Triggers: “reverse engineer”, “catalog features”, “feature inventory”, “code map”, “docs to code mapping”, “binary analysis”.
  • shared — Shared reference documents for multi-agent skills (not directly invocable)
  • standards — Language-specific coding standards and validation rules. Provides Python, Go, Rust, TypeScript, Shell, YAML, JSON, and Markdown standards. Auto-loaded by /vibe, /implement, /doc, /bug-hunt, /complexity based on file types.

  • Single-page catalog All skills on one page — easier to grep or Ctrl-F than browsing by family.

  • Decision Tree "Which skill do I need next?" — single source of truth.

  • Skill API Frontmatter fields, context declarations, enforcement status.

  • Skill Router Routing rules: which skill to use for which task.