Bo

Run multi-judge consensus.

Investigate bugs and root causes.

Dispatch parallel agents.

Coordinate multiple Codex agents.

Run autonomous improvement loops.

Explore and write findings.

'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.'

Generate and validate repo docs.

Review completed work and learn.

Decompose goals into issue plans.

Write compact session handoffs.

Capture a session learning.

Find focused refactor hotspots.

Create or refine PRODUCT.md.

Execute epics through waves.

Validate code readiness.

Run discovery, crank, validation.

Show AgentOps next action.

Trace decisions through artifacts.

Implement one tracked issue.

Run release validation.

Show AgentOps work status.

Separate goals from implementation.

Reverse-engineer product specs.

Recover session context.

Check knowledge flywheel health.

Maintain AgentOps goals.

Explain AgentOps workflows.

Track issues with bd/br, triage with bv, and convert plans to beads.

Repair skill hygiene.

Run repository security scans for vulnerabilities, dependency risk, secrets, and release gates.

Record Brownian Ratchet gates.

Provide repo coding standards.

Convert AgentOps skill formats.

Mine transcripts into learnings.

Draft or improve README docs.

Load relevant .agents context.

Shared AgentOps skill contracts.

Trace artifact provenance.

Research an upstream repo.

Use official OpenAI docs.

Prepare PR commits and body.

Implement a scoped OSS PR.

Validate PR scope and quality.

Run composable security analysis.

Sync AgentOps skills.

Learn from PR outcomes.

Scaffold or audit OSS docs.

Plan an open source PR.