agent-debate

Use when a consequential decision has genuinely competing positions that need bounded adversarial review before a human decision.

SKILL.md

GitHub

agent-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

  1. 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.
  2. Launch two reviewer lanes 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.
  3. 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_cycles limit sets the cap; evaluate the policy before each reviewer spawn.
  4. 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.