mcpField
C
council-review

council-review

github.com/alinaqiTesting & quality71173 installsUpdated 2w ago
Repository

About council-review

A Claude Code skill that defines when to request multi-model validation for plans, architecture changes, and pull requests. It uses installed validation and review commands, with configurable reviewer panels and approval thresholds.

Key features

  1. 01Automatic plan validation
  2. 02Architecture decision reviews
  3. 03Pre-completion PR reviews
  4. 04Configurable reviewer panels
  5. 05Approval thresholds

Use cases

  1. 01Validate a plan before execution
  2. 02Seek independent architecture feedback
  3. 03Review a pull request before completion

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/alinaqi/maggy --skill council-review

After installing, type/council-reviewin Claude Code to confirm it loaded.

Details

From the project README

Council of Experts — Multi-Model Validation

When to Auto-Trigger

Plans (auto_validate_plans)

When you write a plan to ~/.claude/plans/, automatically validate it:

~/bin/validate-plan --threshold 2 ~/.claude/plans/<plan-file>.md
  • 2+ of 3 approve → execute immediately
  • 1 of 3 → surface reviewer feedback to user before proceeding
  • 0 of 3 → revise plan, re-validate
Architecture Decisions (auto_review_architecture)

When making architectural changes (new services, API redesigns, database schema changes), run:

~/bin/review --all "Review this architecture: <summary>"
PR Review (auto_review_prs)

Before marking a PR as done, run:

~/bin/review --all --file <changed-files>

Configuration

Council behavior is configured in ~/.claude/council.yaml. The Maggy dashboard (Settings > Council) also manages this config.

Chief of the Council

chief: claude-fable-5 — Claude Fable 5 (Anthropic's most capable widely-released model, GA 2026-06-09) leads every panel as the chief: it reviews first and casts the deciding synthesis. Invoked via ~/bin/claude-fable-5. Override the chief in ~/.claude/council.yaml.

Reviewer Contexts

The chief leads each context, followed by the panel:

ContextDefault ReviewersWhen
planClaude Fable 5 (chief), DeepSeek Pro, Codex, Gemini ProBefore executing any plan
reviewClaude Fable 5 (chief), DeepSeek Pro, KimiCode review, PR review
architectureClaude Fable 5 (chief), DeepSeek Pro, Gemini Pro, GrokSystem design, schema changes
Threshold Rules

The threshold setting controls how many approvals are needed:

  • threshold: 2 with 3 reviewers → need 2/3 to auto-execute
  • Clamped to [1, reviewer_count] — can't be 0 or exceed available reviewers

Model Inventory

All 13 tiers are listed in ~/.claude/council.yaml under models:. Each has:

  • id — unique identifier
  • cmd — CLI command to invoke (null for Claude models, which are the host)
  • tier — routing priority (0=cheapest, 12=most capable)
  • label — human-readable name

Use POST /api/models/health to verify all models are responding.

How This Skill is Used

This skill is loaded by Claude Code on session start. It provides the behavioral rules for when to invoke multi-model validation. The actual execution happens via ~/bin/validate-plan and ~/bin/review which are already installed.

Do not skip council validation for CLAUDE-tier tasks. The whole point is that architecture and security decisions get independent verification before execution.

Comments

Markdown supported