CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

model-switch

Switch the Manager Agent's own LLM model. Use when the human admin requests changing the Manager's model.

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

88%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.

A strong operational skill: fully executable commands, explicit per-runtime apply and verification steps, and a genuine failure-recovery path with validation before any config mutation. The only weaknesses are mild padding in one explanatory paragraph and a reference table that could be split out of the main file.

DimensionReasoningScore

Conciseness

The body is dense and operational with no explanation of concepts Claude already knows, but the paragraph explaining re-running the script ('because nobody knew its real window') pads a non-obvious point with an illustrative scenario that could be trimmed. This is the score-4 anchor: efficient with minor instances of over-explanation.

4 / 5

Actionability

Guidance is fully executable: copy-paste-ready bash commands with three examples covering the flag variations, the exact script path, per-runtime verification steps (`openclaw gateway restart`, reading back `active_llm.provider_id`), and the exact error string `ERROR: MODEL_NOT_REACHABLE` with concrete Higress Console remediation. This matches the score-5 anchor of copy-paste-ready commands covering common cases.

5 / 5

Workflow Clarity

The multi-step process is clearly sequenced (numbered script behavior, runtime-specific apply/verify steps) with explicit validation checkpoints: a connectivity test before any config change, post-apply model verification, and a feedback loop on failure (error output → admin configures provider/route → retry). This matches the score-5 anchor including error-recovery feedback loops; score 4's 'minor validation gaps' does not apply since both pre-change and post-change validation are present.

5 / 5

Progressive Disclosure

The bundle is a single script referenced clearly in the usage section, and the body is organized into well-labeled sections. The minor gap is that the pre-configured models table is inline reference data that could live in a separate reference file — the score-4 anchor of good structure with minor organization gaps fits; the fully split, reference-heavy structure of score 5 is not present.

4 / 5

Total

18

/

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 concise, well-voiced description that clearly answers both what and when with an explicit 'Use when...' clause and minimal conflict risk. Its weakness is coverage: it names only one action and few trigger-term variations, leaving it borderline against sibling skills phrased with other natural terms like 'update' or 'change your model'.

Suggestions

Enumerate the concrete capabilities beyond 'switch' (e.g., tests model reachability, updates context-window/reasoning settings, applies the change per runtime) to raise specificity.

Add natural trigger variations users might say, such as 'change model', 'update model', or 'use a different LLM', to broaden trigger-term coverage.

Optionally mention the pre-configuration/failure-recovery angle (adding unknown models, provider routing on failure) so the description reflects the skill's full scope.

DimensionReasoningScore

Specificity

The description names the domain ("Switch the Manager Agent's own LLM model") with one concrete action, but does not enumerate several specific actions as the score-4 anchor requires. It matches the score-3 anchor of naming a domain and 1-2 concrete actions without comprehensive coverage.

3 / 5

Completeness

"Switch the Manager Agent's own LLM model" clearly states what the skill does, and "Use when the human admin requests changing the Manager's model" is an explicit, concrete trigger clause. Both what and when are clearly and explicitly answered, matching the score-5 anchor; the 'when' here is fully explicit so the score-4 anchor ('when could be more explicit') does not fit.

5 / 5

Trigger Term Quality

"requests changing the Manager's model" and "Switch" are natural phrases an admin would say, but common variations and synonyms ("update model", "swap model", "different model") are absent. This fits the score-3 anchor of some relevant keywords missing common variations rather than the good-but-incomplete coverage of score 4.

3 / 5

Distinctiveness Conflict Risk

The trigger is narrowly scoped to changing the Manager Agent's own model at an admin's request — a clear niche with distinct triggers and minimal conflict risk with other skills. This matches the score-5 anchor.

5 / 5

Total

16

/

20

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

agentscope-ai/AgentTeams

Path

manager/agent/skills/model-switch/SKILL.md

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.