agentops

Expert code review specialist. Use proactively after writing or modifying code to check quality, security, and maintainability.

Deep codebase exploration and analysis. Use for understanding code architecture, finding patterns, and gathering context before making changes.

Manage PROGRAM.md/AUTODEV.md contracts for bounded autonomous development loops.

Use bd issue tracking: create, update, route, and close dependency-aware beads.

Initialize AgentOps project files, goals, product docs, README, hooks, and .agents state.

Separate goals from implementation; explore options and capture design decisions.

Investigate bugs or audit code with repro evidence, root cause analysis, and fixes.

Coordinate 2+ Codex agents on bounded parallel tasks with file ownership.

Compile .agents knowledge into .agents/compiled wiki and run maintenance checks.

Analyze code complexity and identify focused refactor targets.

Convert AgentOps skills between Codex, Cursor, and raw bundle formats.

Run multi-judge consensus for validation, brainstorming, or research.

Execute an epic hands-free through waves until work is done or blocked.

Audit, update, and validate dependencies, vulnerabilities, and licenses.

Validate product fit before discovery using PRODUCT.md and scope checks.

Orchestrate brainstorm, research, plan, and pre-mortem into an execution packet.

Generate, sync, and validate docs against repo evidence.

Run or review private overnight AgentOps compounding sessions.

Run autonomous improvement: goals, planning, implementation, validation, repeat.

Check knowledge flywheel health, velocity, staleness, and pool depth.

Mine transcripts and artifacts into reusable learnings, decisions, and patterns.

Maintain GOALS.yaml and GOALS.md fitness specs, directives, and drift.

Improve OpenShift/Kubernetes Grafana dashboards and validate PromQL/GitOps changes.

Create compact session handoffs for continuation after pause or compaction.

Consolidate knowledge across .agents directories and promote useful items.

Repair skill hygiene: frontmatter, links, references, manifests, and parity drift.

Implement one tracked issue with focused edits, tests, and closure evidence.

Load relevant .agents knowledge into the current session context.

Turn a mature .agents corpus into packets, belief books, briefings, and gaps.

Build an external-knowledge wiki from clipped articles, papers, and transcripts.

Use official OpenAI docs for API/product questions needing current citations.

Scaffold or audit open source docs such as README, CONTRIBUTING, and changelog.

Profile, benchmark, and optimize performance regressions or hotspots.

Decompose goals into issue-ready plans, waves, dependencies, and validation checks.

Review completed work, capture learnings, and feed follow-up knowledge.

Implement a scoped open source PR from a researched contribution plan.

Plan an open source PR with scope, acceptance criteria, and risk.

Prepare PR commits, validation evidence, and review-ready PR body.

Research an upstream repo before contributing: rules, patterns, and expectations.

Learn from PR outcomes and update contribution lessons.

Validate PR scope, isolation, quality, and upstream alignment.

Stress-test a plan before implementation with council or inline review.

Create or refine PRODUCT.md through mission, persona, and value prompts.

Trace where knowledge, decisions, or artifacts came from.

Run gates, commit intentionally, push the branch, and confirm remote state.

Detect AgentOps setup and give one concise next action.

Record and verify Brownian Ratchet progress gates.

Draft or improve a project README using structured evidence and review.

Recover session context after compaction from RPI, evolve, and recent artifacts.

Probe docs or skills with adversarial persona tests and consolidate findings.

Plan and execute safe refactors with regression checks at each step.

Run release validation, changelog/version steps, tagging, and release notes.

Explore a codebase or question and write evidence-backed findings.

Capture a quick learning or session note for later compounding.

Reverse-engineer a product into feature catalog, code map, and specs.

Review code, PRs, diffs, or agent output for correctness and risk.

Run full lifecycle: discovery, crank implementation, validation, and report.

Create project, feature, component, or boilerplate scaffolds.

Author and manage holdout behavioral validation scenarios.

Run composable security analysis across binaries, prompts, traces, and policies.

Run repository security scans and release-gating checks.

Shared reference contracts used by other AgentOps skills.

Provide language and repo standards for implementation, docs, tests, and review.

Show current work, recent validation, flywheel health, and next action.

Dispatch isolated agents for parallel tasks with file ownership and evidence.

Generate tests, analyze coverage, and plan TDD validation.

Trace decisions and concepts through history, artifacts, and git.

Reinstall or sync AgentOps skills from the latest source.

Explain AgentOps operating model, lifecycle, skills, hooks, and context.

Run post-implementation validation: vibe, post-mortem, retro, and forge.

Validate code readiness with complexity checks and council or inline review.

28 hooks across 7 events