Review and improve architecture specs, ADRs, plugin or agent directory proposals, and other design documents by running multiple independent AI reviewers such as Claude, Qoder, Codex, or Cursor Agent against the same artifact, normalizing P1/P2/P3 findings, and iterating only on blocking or high-risk issues.
SkillScore72Maintained
Automatic
◆Product Management pack
Last verified 8/13/2026
What it changes
Passed 14 of 20 test cases with the skill, and 9 of 20 without it.
Without skill44%
With skill72%
Checks it passed that the model alone missed
runs git diff --check during local validation
expects the triad review script to output normalized json
executes scripts/run-triad-review.mjs with the --target flag
and 4 more in the full report
7 fixed11 passed either way7 still failing0 broke
Pass-rate lift
+28 pp
Tokens
-3%
Agent turns
0%
Response time
-47%
Model gemini-3.6-flashMeasured 8/13/2026v118 of 25 cases passed with skill · 20 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 7 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 7
Prompt
What is the very first step the lead agent must take when initiated with a spec review request for `docs/specs/search.md`?
What was checked, without and with the skill
Checked
Without skill
With skill
resolves the target spec path and acceptance gate as the first workflow step
Missed
Passed
case-25 · gemini-3.6-flash · 8/13/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md42 lines3 filesincludes scripts
Triangulate Spec Review
Use independent AI reviewers as an evaluation surface for specs. The lead agent
owns synthesis, edits, validation, and the final recommendation.
Workflow
Resolve the target spec path and the acceptance gate. If the user did not
name dimensions, default to complexity, convenience, and evolution.
Read the target spec and nearby repo instructions. Do not pass your suspected
fixes or conclusions to reviewers.
Use the same read-only prompt for every reviewer. Ask for structured
P1/P2/P3 findings and a p1_p2_clear boolean.
Run at least two reviewers; prefer three when available. Use
scripts/run-triad-review.mjs for repeatable local runs.
Normalize findings. Treat reviewers as evidence, not authority:
fix convergent P1/P2 issues first, challenge weak or contradictory
findings, and leave P3 as backlog unless it is cheap and clarifying.
If the user asked for edits, patch only the owning spec or directly related
helper docs. Keep unrelated refactors out of the review loop.
Run local validation such as git diff --check, relevant tests, and targeted
rg checks for renamed concepts or stale paths.
Repeat the reviewer pass until every requested reviewer reports no P1 or
P2, or until the user stops the loop.
Resources
Read references/review-loop.md for the prompt contract, severity rubric,
command matrix, and iteration patterns.
Run this skill's scripts/run-triad-review.mjs --target <path> script,
resolved relative to the triangulate-spec-review skill directory, to execute
a read-only review round and write normalized JSON output.
Guardrails
Pass raw artifacts and task-local context to reviewers, not intended answers.
Keep every reviewer prompt materially identical unless a tool requires command
syntax changes.
Do not let reviewers edit files. The lead agent applies changes after
comparing findings.