council

Spawn parallel task agents to explore a given area of the codebase from multiple angles, then use their findings to answer the question or build a plan. Use when a task needs broad fan-out exploration across many files before acting.

Quality

68%

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 ./.agents/skills/council/SKILL.md

Quality

Content

61%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 body is admirably token-efficient and lays out a recognizable three-phase workflow with a plan-mode branch, but it is light on actionable specifics: how to phrase sub-agent prompts, how to vary them 'out of the box', and how to synthesize ten sets of findings are all left implicit. It reads as a skeleton that relies on Claude improvising the hard parts.

Suggestions

Add one concrete example of a task-agent prompt (e.g. a template with the area of interest inserted) so the spawn step is executable rather than implied.

Replace 'use the information to do what the user wants' with a short synthesis step — e.g. 'reconcile overlapping findings, note contradictions between agents, and summarize the consensus before answering'.

Specify what 'out of the box for variance' means concretely (2-3 example angle variations), and add a checkpoint after the agents finish (e.g. verify findings are grounded in real files before building the plan).

Dimension	Reasoning	Score
Conciseness	The 7-line body is lean and assumes Claude's competence with zero padding or explanation of known concepts, matching 'every token earns its place'. Not 4 because there is no over-explanation to trim — only trivial stylistic repetition; the body stays well under budget.	5 / 5
Actionability	Guidance is high-level — 'Dig around the codebase', 'gather general information such as keywords and architecture overview', 'use the information to do what the user wants' — with no concrete commands, example sub-agent prompts, or synthesis method; the only specific parameters are 'n=10' and 'out of the box for variance'. Matches 'minimal concrete guidance; high-level hints but missing specific steps'. Not 3 because nothing executable or copy-paste-ready is provided; not 1 because a concrete agent count and a rough step structure do exist.	2 / 5
Workflow Clarity	A three-step sequence is listed (scout → spawn agents → apply findings) plus a plan-mode conditional branch, but step 3 ('use the information to do what the user wants') is undefined and there are no synthesis or validation checkpoints for the agents' results — and spawning 10 agents is a batch operation, capping this dimension. Matches 'steps listed but validation gaps; checkpoints missing or implicit'. Not 4 because no checkpoints exist for reconciling agent output; not 2 because the steps are numbered, ordered, and mode-aware rather than incoherent.	3 / 5
Progressive Disclosure	No bundle files exist, and the sub-50-line body inlines nothing that belongs in a separate file, so nothing is buried or misplaced. Not 5 because the body has no organizational structure beyond the flat numbered list — the plan-mode note is not signposted as its own section; not 3 because there is no misplacement or navigation problem at this size.	4 / 5
	Total	14 / 20 Passed

Description

75%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.

A solid description: third-person voice, explicit what-and-when, concrete mechanics, and decent trigger terms. Its main weakness is that the 'when' clause is phrased in design vocabulary ('broad fan-out exploration') rather than phrases a user would naturally say, slightly limiting trigger matching and completeness.

Suggestions

Rewrite the 'Use when' clause with user-voice triggers, e.g. 'Use when the user asks to explore, investigate, or research an unfamiliar area of the codebase, or wants a broad survey before acting'.

Add common synonyms/extensions users would say — 'search the codebase', 'survey', 'understand an area' — to improve trigger_term_quality toward comprehensive coverage.

Consider enumerating one or two more downstream uses of the findings (e.g. summarize, produce a report, answer architecture questions) to lift specificity.

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete actions — 'Spawn parallel task agents to explore a given area of the codebase from multiple angles, then use their findings to answer the question or build a plan' — matching the 'several specific actions, minor gaps' anchor. Not 5 because coverage of downstream uses is partial (answer/build plan but no other outcomes); not 3 because it goes beyond 1-2 actions with specific mechanics (parallel agents, multiple angles, findings synthesis).	4 / 5
Completeness	Both 'what' (spawn parallel agents, explore from multiple angles, use findings to answer or plan) and an explicit 'when' ('Use when a task needs broad fan-out exploration across many files before acting') are present. Not 5 because the when-clause uses internal task-jargon rather than concrete user-voice trigger phrases; not 3 because the 'when' is explicit, not missing or weakly implied.	4 / 5
Trigger Term Quality	Includes natural, sayable terms — 'parallel task agents', 'explore', 'codebase', 'fan-out exploration', 'many files' — giving good keyword coverage. Not 5 because common variations users would actually say ('search', 'investigate', 'survey', 'understand this area') are missing; not 3 because multiple relevant natural terms are present, not just one or two generic ones.	4 / 5
Distinctiveness Conflict Risk	The parallel-agent fan-out niche is mostly distinct from typical single-skill domains, with distinctive triggers ('parallel task agents', 'fan-out exploration'). Not 5 because it can overlap with generic codebase exploration/search/research skills; not 3 because the multi-angle agent orchestration framing sets it apart from ordinary file-format or task skills.	4 / 5
	Total	16 / 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: simstudioai/sim
Path: .agents/skills/council/SKILL.md
Commit: 8725250

Reviewed: 3 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.