agent-verify
Use fresh subagents to verify new work, compare before/after improvements, or improve a prompt or skill step by step in red-green-refactor rounds, in code, skills, prompts, documents, or agent behavior. Inspect real artifacts and test results with independent QA, matched trials, and regression checks.
Pinned to revision afb03e54d58b, so it is the text this page describes rather than whatever the author pushed since.
Files
- skills/agent-verify/SKILL.md
- skills/agent-verify/agents/openai.yaml
- skills/agent-verify/evals/2026-10-31-comparison.md
- skills/agent-verify/evals/2026-10-31-installation.md
- skills/agent-verify/evals/2026-10-31-subagent-verification.md
- skills/agent-verify/evals/2026-11-05-improvement-loop.md
- skills/agent-verify/evals/README.md
- skills/agent-verify/evals/fixtures/loop-forward/diffs/01-retry.diff
- skills/agent-verify/evals/fixtures/loop-forward/diffs/02-typo.diff
- skills/agent-verify/evals/fixtures/loop-forward/diffs/03-remove-endpoint.diff
- skills/agent-verify/evals/fixtures/loop-forward/diffs/04-config-and-test.diff
- skills/agent-verify/evals/fixtures/loop-forward/diffs/05-fix-null.diff
- skills/agent-verify/evals/fixtures/loop-forward/final-prompt-run2.md
- skills/agent-verify/evals/fixtures/loop-forward/prompt.md
- skills/agent-verify/evals/fixtures/loop-forward/request.md
- skills/agent-verify/evals/fixtures/loop-forward/run1-ledger.json
- skills/agent-verify/evals/fixtures/loop-forward/run2-ledger.json
- skills/agent-verify/evals/fixtures/loop-ledger.json
- skills/agent-verify/evals/fixtures/loop-no-red.json
- skills/agent-verify/evals/fixtures/loop-premature-green.json
- skills/agent-verify/evals/fixtures/sample-output.txt
- skills/agent-verify/evals/fixtures/tampered-receipt.json
- skills/agent-verify/evals/fixtures/valid-receipt.json
- skills/agent-verify/references/comparison.md
- skills/agent-verify/references/evidence.md
- skills/agent-verify/references/improvement-loop.md
- skills/agent-verify/references/improvement.md
- skills/agent-verify/references/installation.md
- skills/agent-verify/references/model-providers.md
- skills/agent-verify/references/progressive-audit.md
- skills/agent-verify/references/qa.md
- skills/agent-verify/references/research-basis.md
- skills/agent-verify/scripts/check_receipt.py
- skills/agent-verify/scripts/score_loop.py
Every link opens the file at its source, pinned to the revision this page describes.