Menu
Council review reference data — expertise weights, structured response format, scoring thresholds, and false positive taxonomy. Background knowledge for council subagents.
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Baseado na classificação ocupacional SOC
Validate plugin SKILL.md frontmatter and audit hook scripts for silent failures. Run validation to check all plugins pass schema, source path, and frontmatter checks. Run hook audit to detect unhandled errors in shell and Python scripts.
Multi-agent development workflow using Agent Teams. Supports four modes: plan (architect + PM debate → plan.md), dev (developer + tester + reviewer iterate → code), full (plan → approval gate → dev), and auto (plan → dev, no gate). Use when tasks benefit from collaborative agent roles with peer messaging.
Consult external AI council (Gemini, Codex, Qwen, GLM-4.7) for thorough reviews and consensus-driven decisions. Use ONLY when explicitly invoked with "/council" or when user says "consult the council", "invoke council", or "council review". Do NOT auto-trigger on generic phrases like "thorough review".
Interactive issue creation optimized for LLM agent teams. Guides users through type-specific question flows (bug, feature, epic, refactor, new project, chore, research spike) and produces structured, machine-parseable GitHub issues that AI coding agents can execute autonomously. Triggers: create issue, plan work, new task, project manager, write ticket, draft issue, plan feature, plan project, start project, create ticket, pm.
| name | council-reference |
| description | Council review reference data — expertise weights, structured response format, scoring thresholds, and false positive taxonomy. Background knowledge for council subagents. |
| user-invocable | false |
| disable-model-invocation | true |
Every council agent MUST return this JSON structure:
{
"consultant": "gemini|codex|qwen|glm|kimi|claude-security|claude-bugs|claude-compliance|claude-history|claude-quality",
"success": true,
"fallback": false,
"confidence": 0.0-1.0,
"severity": "critical|high|medium|low|none",
"findings": [
{
"type": "security|performance|quality|architecture|bug",
"severity": "critical|high|medium|low",
"description": "...",
"location": "file:line",
"recommendation": "..."
}
],
"summary": "One-paragraph summary"
}
location: MANDATORY for /council review findings. Format: file:line (e.g. src/api.ts:42).
Task Gemini Codex Qwen GLM-4.7 Kimi K2.5
Security 0.90 0.80 0.70 0.75 0.70
PR Review 0.85 0.90 0.80 0.75 0.80
Architecture 0.85 0.70 0.65 0.80 0.75
Code Quality 0.70 0.80 0.90 0.70 0.80
Performance 0.75 0.85 0.85 0.70 0.80
Brainstorm 0.65 0.60 0.90 0.85 0.80
Algorithms 0.70 0.75 0.85 0.85 0.80
Debugging 0.75 0.90 0.80 0.75 0.80
Score Meaning
───── ───────────────────────────────────────────────────────
0 False positive. Doesn't hold up to scrutiny.
25 Might be real, but unverified. Could be false positive.
50 Real but minor. Unlikely to occur in practice.
75 Verified real. Will impact functionality. Important.
100 Confirmed. Frequent in practice. Evidence conclusive.
───── ───────────────────────────────────────────────────────
Filter threshold: Only findings scoring >= 80 appear in the final report.
Score 0 (auto-reject) if a finding matches any of these:
Weighted Score = Σ(Opinion × Expertise × Confidence) / Σ(Expertise × Confidence)