What problem does it solve? Debugging complex bugs often suffers from single-pass reasoning that misses root causes or picks risky fixes. This Skill orchestrates a structured multi-agent debate where N independent solvers argue over R rounds and an isolated judge synthesizes one auditable recommendation for a registered bug. ## Core Features & Use Cases - Three Debate Modes: diagnosis (competing causal hypotheses), repair (competing fix strategies), and spec (recommendation on whether code, tests, or spec is correct). - Isolated Judge with Frozen Rubric: the judge sees only anonymized, shuffled final proposals and applies a rubric frozen before the debate, preventing echo chambers and reward hacking. - Auditable State on Disk: every round, convergence metric, and the final synthesis is written under the bug's debate folder, with explicit quorum, timeout, and failure handling. - Optional External Harnesses: can include CLI agents like Codex or Gemini as debaters or critics, only with explicit user consent and never for restricted-visibility bugs. - Use Case: A registered bug has conflicting evidence about its root cause. Run the debate in diagnosis mode with 3 solvers and 2 rounds, then receive one judge-synthesized hypothesis with a discriminating probe to confirm it. ## Quick Start Ask the agent to run a multi-agent debate on bug BUG-20260715-A7K3 in diagnosis mode with the default 3 solvers and 2 rounds.
