Run a long or multi-session task to completion with a written brief, acceptance checks, a progress file that survives context resets, an evidence review and a resume prompt. Use when the user asks to finish, carry through or resume a long task, or invokes…
Compare agent harnesses (Claude Code, Codex, Gemini CLI, OpenCode through allies) on the same frozen tasks, judged blind, with score, pass rate, cost and the tasks where they differ - real data for routing work to allies. Use when the user asks which agent or…
Explain what MOAA installed and how to use it (council, arena, War Room, allies, strata, skill catalog, installer commands). Use when the user asks what MOAA is, what it can do, how to start, or which MOAA tool fits a task.
Assemble task-specific subagents with distinct working tendencies. Use for parallel specialist teams, independent competing solutions, multi-model comparisons, or a council that evaluates and combines results. Choose split, ensemble, or hybrid execution and…
Build and maintain a Karpathy-style LLM Wiki - compile papers, notes and project sources once into linked Markdown pages (index, log, schema) that Claude writes and keeps current, answer from the pages instead of re-reading raw sources, and lint the link…
Improve something against a fixed number - edit, measure with a script the agent may not change, keep the change if the number improved, otherwise revert, repeat until the budget runs out (after Karpathy's autoresearch). Use for tuning code speed, model loss,…
Measure whether a council of subagents beats one agent loop on your own tasks, per unit of cost - the same frozen tasks run solo and as a council, judged blind, with the tasks whose outcome flips. Use when the user asks whether subagents or the council are…
Record findings as a typed graph (claims, sources, decisions, with supports, cites, contradicts and depends_on edges) that parallel agents share, and verify with a deterministic grounding check that every claim traces to a source. Use for multi-agent…