llm-council
Orchestrate a configurable, multi-member CLI planning council (Codex, Claude Code, Gemini, OpenCode, or custom) to produce independent implementation plans, anonymize and randomize them, then judge and merge into one final plan. Use when you need a robust, bias-resistant planning workflow, structured JSON outputs, retries, and failure handling across multiple CLI agents.
83
78%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/llm-council/SKILL.mdQuality
Discovery
85%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong description that clearly articulates a unique, specific workflow involving multi-agent planning councils with anonymization and merging. It includes both 'what' and 'when' clauses with explicit triggers. The main weakness is that trigger terms are somewhat specialized and may not fully cover the natural language variations a user might employ when seeking this capability.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: orchestrate a multi-member CLI planning council, produce independent implementation plans, anonymize and randomize them, judge and merge into one final plan. Also mentions structured JSON outputs, retries, and failure handling. | 3 / 3 |
Completeness | Clearly answers both 'what' (orchestrate a configurable multi-member CLI planning council to produce, anonymize, judge, and merge plans) and 'when' (explicit 'Use when you need a robust, bias-resistant planning workflow, structured JSON outputs, retries, and failure handling across multiple CLI agents'). | 3 / 3 |
Trigger Term Quality | Includes some relevant keywords like 'planning council', 'Codex', 'Claude Code', 'Gemini', 'OpenCode', 'implementation plans', and 'CLI agents'. However, these are fairly specialized terms; a user might more naturally say 'plan comparison', 'multi-agent planning', or 'council of experts' — some common variations are missing. | 2 / 3 |
Distinctiveness Conflict Risk | Highly distinctive — the concept of a multi-member CLI planning council with anonymization, randomization, and merging is a very specific niche. It names specific tools (Codex, Claude Code, Gemini, OpenCode) and a unique workflow pattern, making it unlikely to conflict with other skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
72%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a solid skill that provides actionable, well-structured guidance for orchestrating a multi-agent planning council. Its strengths are concrete CLI commands, a detailed JSON configuration example, and good progressive disclosure to reference files. Weaknesses include some redundancy between sections (intake questions, session management constraints stated twice) and workflow validation steps that could be more explicit about error conditions and recovery.
Suggestions
Remove duplicated guidance: consolidate intake question instructions (Quick Start + Workflow) and session management rules (Workflow step 7 + Constraints) into single locations.
Make the retry/failure handling in Workflow step 4 more explicit—specify what validation checks are performed, what constitutes a retryable vs. fatal failure, and the exact retry command.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is mostly efficient but has some redundancy—session management instructions are repeated in both the Workflow and Constraints sections, and the intake question guidance appears in both Quick Start and Workflow. Some phrasing could be tightened (e.g., 'Tell the user that answering intake questions is optional' is repeated). | 2 / 3 |
Actionability | Provides concrete CLI commands (`python3 scripts/llm_council.py run --spec /path/to/spec.json`, `./setup.sh`), a complete JSON configuration example with multiple agent types, and specific file paths for outputs. The guidance is specific and executable rather than abstract. | 3 / 3 |
Workflow Clarity | The 7-step workflow is clearly sequenced and includes retry logic (step 4) and validation of Markdown structure. However, the validation/retry step is somewhat vague ('retry up to 2 times on failure' without specifying what constitutes failure or how to retry), and the session management step (7) is more of an operational constraint than a workflow step. Missing explicit validation checkpoints between judging and final plan generation. | 2 / 3 |
Progressive Disclosure | The skill has a clear overview structure with well-signaled one-level-deep references to architecture, prompt templates, plan templates, and CLI notes. Content is appropriately split between the main skill file and reference documents, with clear navigation paths. | 3 / 3 |
Total | 10 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- am-will/codex-skills
- Commit
ca2c3b3
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.