agent-debate
Use when a consequential decision has genuinely competing positions that need bounded adversarial review before a human decision.
SKILL.mdskills/agent-debate/SKILL.md
GitHubagent-debate
A bounded adversarial review of one consequential trade-off. Debate makes the evidence visible to the decision maker; it does not settle truth by vote and does not authorize implementation.
Use only when
- both options are plausible and materially different;
- the decision is consequential enough to justify two independent reviewers;
- identifiable evidence could inform or falsify each side.
Do not debate a factual question that a direct check can settle, a naming preference, or a decision an authorized human already made.
Setup
- Write one evidence packet: the decision and who owns it, the two positions as testable statements, shared evidence paths, and the criteria for judging. Template: references/debate_protocol.md.
- Launch two
reviewerlanes in parallel, one position each, both on a strong model tier, both read-only, both with the same packet. Each argues the strongest form of its side and cites evidence. - State the round cap before launch and stop there. A rebuttal round, if
the cap allows one, answers the other side's strongest point. In harness
mode the policy's
max_cycleslimit sets the cap; evaluate the policy before each reviewer spawn. - A reviewer that changes its conclusion says so plainly; do not simulate disagreement.
Synthesis
The primary-agent, or a separate synthesizer with no search tools, writes:
- agreed facts, with evidence;
- contested claims, with the evidence on each side;
- missing or failed reviewer outputs (recorded, never invented);
- a recommendation, how uncertain it is, and what finding would change it;
- the pending human decision: approve, decline, or revise.
Nothing is implemented and no canonical memory changes until the human
decides. If the outcome is worth keeping, write a memory proposal as described
in agent-context-budget.
Prohibitions
- No majority vote as evidence or acceptance.
- No silent retry, model switch, context discard, or invented missing side.
- No source edits, commits, merges, or pushes during the debate.
- No direct write to canonical memory.