Run a consensus-style multi-agent review of a PR with severity-based findings.
SkillScore59Maintained
Automatic
◆Coding pack
Last verified 8/20/2026
What it changes
Passed 13 of 22 test cases with the skill, and 12 of 22 without it.
Without skill55%
With skill59%
Checks it passed that the model alone missed
Assigns CRITICAL severity rating to hardcoded API key and secret findings.
Includes a dedicated consensus summary section with a verdict of APPROVE or REQUEST_CHANGES.
2 fixed11 passed either way8 still failing1 broke
Pass-rate lift
+5 pp
Tokens
-14%
Agent turns
0%
Response time
-22%
Model gemini-3.6-flashMeasured 8/20/2026v113 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 2 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 2
Prompt
Run a multi-agent consensus review on PR #990 which updates environment config files. Check for hardcoded credentials and assign severity.
What was checked, without and with the skill
Checked
Without skill
With skill
Assigns CRITICAL severity rating to hardcoded API key and secret findings.
Missed
Passed
case-16 · gemini-3.6-flash · 8/20/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md33 lines8 filesincludes scripts
multi-pr-review
Purpose
Run a consensus-style multi-agent review of a PR with severity-based findings.
Routing
Use when: Use when the user asks to run a consensus-style multi-agent review of a PR with severity-based findings.
Do not use when: Do not use when the request is asking for planning documents, high-level strategy, or non-executable discussion; use the relevant planning or design workflow instead.
Outputs: Outcome from multi-pr-review: task-specific result plus concrete action notes.
Success criteria: Returns concrete actions and decisions matching the requested task, with no fabricated tool-side behavior.
Trigger Examples
Positive
Use the multi-pr-review skill for this request.
Help me with multi-pr-review.
Use when the user asks to run a consensus-style multi-agent review of a PR with severity-based findings.
multi-pr-review: provide an actionable result.
Negative
Do not use when the request is asking for planning documents, high-level strategy, or non-executable discussion; use the relevant planning or design workflow instead.
Do not use multi-pr-review for unrelated requests.
This request is outside multi-pr-review scope.
This is conceptual discussion only; no tool workflow is needed.
Runtime Prompt
Current runtime prompt length: 872 characters.
Runtime prompt is defined directly in ../multi-pr-review.json.