bird
01Command
Domain Authority & Final Arbiter — defines what is correct vs merely working.
Seven legends. One coach. Zero groupthink. Put any agent on any provider — Claude, Codex, Gemini, or Ollama — for both evals and live /team sessions, each on your own subscription.
Command
Domain Authority & Final Arbiter — defines what is correct vs merely working.
Command
Code review a pull request (local only — never posts to GitHub).
Command
Deletion-Bias Enforcer — finds duplication, over-engineering, and maintenance debt.
Command
Dream Team eval runner — run agent evaluation scenarios, auto-score with Coach K, and open the web app for human review.
Command
Quality & Risk Enforcer — finds edge cases, race conditions, and hidden assumptions.
Command
End-of-session learning loop — analyze signals, approve instincts, capture directives, refresh MEMORY.md. Invokes dreamteam learn; one source of behavioral truth shared with the team.md SESSION LEARNING step.
Command
Context Synthesizer & Team Glue — synthesizes outputs, produces summaries and ADRs.
Command
Strategic Systems Architect — designs clean system boundaries and anticipates second-order effects.
Command
Stability, Integration & Defense — ensures components work together and stay working in production.
Command
Primary Code Executor — turns specs into production-ready code with tests.
Command
Dream Team orchestration — solve problems with a full AI agent team (Coach K coordinates MJ, Bird, Shaq, Kobe, Pippen, Magic, and Drexler).
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: