llm-council

Query multiple LLM models in parallel from CodeAct and cross-reference their responses

Quality

71%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/llm-council/SKILL.md

Quality

Content

82%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The skill body is highly actionable with executable code throughout and a clear synthesis workflow. It is efficient and well-organized, with only minor conciseness and validation-checkpoint gaps holding it back from top marks.

Suggestions

Add an explicit validation/retry checkpoint in the batch flow (e.g., inspect responses for 'Error: ...' slots and retry or substitute a fallback model before synthesis) to strengthen workflow_clarity for this batch operation.

Consolidate the repeated per-backend caveats (the table, the per-backend line-up notes, and the Notes section restate the Bedrock/no-override limitation) into a single location to tighten conciseness.

Consider moving the full multi-line backend line-ups into a short reference or condensing them, so the core council pattern stays prominent in SKILL.md.

Dimension	Reasoning	Score
Conciseness	The body is largely efficient — it skips explaining what LLMs are and jumps straight to backend behavior, line-ups, and executable snippets — with only minor instances of restated guidance that could be trimmed (e.g., backend caveats repeated across the table, line-up, and Notes).	4 / 5
Actionability	It provides fully executable, copy-paste-ready code for the single-call, parallel council, many-prompts, and full synthesis example, covering the common cases with concrete model lists and expected result shapes.	5 / 5
Workflow Clarity	The 'After collecting responses' section gives a clear 4-step synthesis sequence, and Notes cover error handling, but there is no explicit validation checkpoint for confirming batch integrity or retrying failed slots, leaving a minor validation gap.	4 / 5
Progressive Disclosure	The content is well-organized into clearly labeled sections (When to use, Default line-up, API, After collecting responses, Full example, Notes) with no external bundle files needed and no deep nested references; it is slightly dense for a single SKILL.md but appropriately structured.	4 / 5
	Total	17 / 20 Passed

Description

61%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is concise and clearly states the skill's purpose, with a strong set of natural trigger keywords. It lacks an explicit 'when to use' clause and only names one concrete action, which limits its specificity and completeness.

Suggestions

Add an explicit 'Use when...' clause (e.g., 'Use when the user wants diverse perspectives, second opinions, or to cross-reference multiple models on one prompt') to satisfy the completeness dimension.

List one or two more concrete actions (e.g., 'synthesize consensus, flag disagreements, and cite which model contributed each insight') to raise specificity.

Tighten the third-person voice by keeping the description purely descriptive (it already is), but consider adding the activation trigger phrasing into the description for completeness.

Dimension	Reasoning	Score
Specificity	The description names the domain (querying multiple LLM models in parallel from CodeAct) and one concrete action (cross-reference their responses), but does not enumerate several specific actions like the score-4/5 anchors expect.	3 / 5
Completeness	It clearly states 'what' the skill does (query multiple LLM models in parallel and cross-reference) but there is no explicit 'Use when...' clause or equivalent trigger guidance, which caps completeness at 3 per the rubric.	3 / 5
Trigger Term Quality	It uses natural phrases users would say ('compare models', 'cross-reference', 'multiple models', 'second opinion', 'vote') though it omits some common variations a user might naturally voice.	4 / 5
Distinctiveness Conflict Risk	The niche is fairly distinct — running parallel multi-model LLM queries from CodeAct — with limited overlap risk, though it could still brush against general LLM-querying skills.	4 / 5
	Total	14 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: nearai/ironclaw
Path: skills/llm-council/SKILL.md
Commit: 2831516

Reviewed: 8 days ago

Table of Contents

Discovery Implementation Validation

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.