worker-model-switch

Switch a Worker Agent's LLM model via agt CLI. Use when the human admin requests changing a Worker's model.

Quality

86%

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

Quality

Content

87%Weight 40%Scale 1-3

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

The body is lean, actionable, and well-structured for a simple single-command skill, with executable examples and clear success/failure behavior. The only notable gap is the absence of a validation or verification checkpoint for the state-changing container recreation.

Suggestions

Add a verification step after the command — e.g., check the CLI prints `worker/<name> configured` and optionally confirm the worker is running the new model — to establish a validate→fix→retry feedback loop for the state-changing recreation.

Include a brief recovery hint in the failure section (e.g., retry after confirming the controller API is reachable) so the workflow loops rather than only reporting the error to the admin.

Clarify that a failed switch leaves the worker on its previous model, so the admin knows the safe state on partial failure.

Dimension	Reasoning	Score
Conciseness	The 40-line body is lean with no explanations of concepts Claude already knows (it never defines LLM, CLI, or container), and each section — command, examples, what-happens, success/failure, edge case — earns its place. Not a 2 because there is no unnecessary explanation or padding to trim.	3 / 3
Actionability	Provides the fully executable command `agt update worker --name <WORKER_NAME> --model <MODEL_ID>` plus two concrete copy-paste examples and explicit success/failure signals. Not a 2 because the guidance is complete and copy-paste ready rather than pseudocode or missing key details.	3 / 3
Workflow Clarity	The single action is unambiguous and supported by numbered 'What happens' steps and on-success/on-failure sections, but a model switch recreates a container (a state-changing operation) with no validate→fix→retry checkpoint or verification that the new model took effect. Per the rubric, missing validation for destructive/state-changing operations caps this at 2; not a 1 because steps and sequence are clearly present.	2 / 3
Progressive Disclosure	At under 50 lines with no bundle files, the well-organized sections (Usage, What happens, On success, On failure, Unknown models) satisfy the rubric's simple-skill allowance for a top score without external references. Not a 2 because organization is clear and content is appropriately inline for its size.	3 / 3
	Total	11 / 12 Passed

Description

85%Weight 40%Scale 1-3

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, specific, and supplies an explicit 'Use when...' trigger that answers both what and when. Its main weakness is limited keyword variation, which keeps trigger term quality at a mid score rather than full marks.

Dimension	Reasoning	Score
Specificity	Quotes concrete action with domain, tool, and verb — "Switch a Worker Agent's LLM model via agt CLI" — naming the specific capability and tool rather than vague language. Not a 2 because it ties a concrete action to a named tool and domain, exceeding the 'names domain and some actions' anchor.	3 / 3
Completeness	Explicitly answers both what ("Switch a Worker Agent's LLM model via agt CLI") and when ("Use when the human admin requests changing a Worker's model"). Not a 2 because the 'Use when...' trigger is explicit rather than merely implied.	3 / 3
Trigger Term Quality	Includes natural phrases like "when the human admin requests changing a Worker's model" that a user would plausibly say, but offers limited variation and no common synonyms (e.g. 'update model', 'swap model'). Not a 3 because coverage of natural terms is thin rather than comprehensive.	2 / 3
Distinctiveness Conflict Risk	Targets a narrow niche (switching a Worker Agent's model via agt CLI) with a distinct admin-driven trigger, making it unlikely to fire for unrelated skills. Not a 2 because the niche and trigger are specific enough to avoid meaningful overlap.	3 / 3
	Total	11 / 12 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: agentscope-ai/AgentTeams
Path: manager/agent/skills/worker-model-switch/SKILL.md
Commit: 785c2db

Reviewed: 6 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.