cursor-model-selection
Configure and select AI models in Cursor for Chat, Composer, and Agent mode. Triggers on "cursor model", "cursor gpt", "cursor claude", "change cursor model", "cursor ai model", "cursor auto mode".
54
62%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./plugins/saas-packs/cursor-pack/skills/cursor-model-selection/SKILL.mdQuality
Discovery
89%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 solid description with excellent trigger terms and clear completeness, explicitly stating both what the skill does and when to use it. The main weakness is that the 'what' portion could be more specific about the concrete actions available (e.g., listing models, setting defaults, comparing capabilities). Overall it performs well for skill selection purposes.
Suggestions
Expand the capability description with more specific actions, e.g., 'Configure and select AI models in Cursor, set default models for Chat/Composer/Agent modes, compare model capabilities, and troubleshoot model availability.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names the domain (AI models in Cursor) and some actions ('configure and select'), but doesn't list specific concrete actions like adding models, switching between models, setting default models, or configuring API keys. | 2 / 3 |
Completeness | Clearly answers both 'what' (configure and select AI models in Cursor for Chat, Composer, and Agent mode) and 'when' (explicit trigger terms listed with 'Triggers on' clause), satisfying the requirement for explicit trigger guidance. | 3 / 3 |
Trigger Term Quality | Includes a strong set of natural trigger terms that users would actually say: 'cursor model', 'cursor gpt', 'cursor claude', 'change cursor model', 'cursor ai model', 'cursor auto mode'. These cover common variations well. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with a clear niche — specifically about AI model configuration within Cursor IDE. The trigger terms are specific enough (combining 'cursor' with model-related terms) to avoid conflicts with general AI or general Cursor skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
35%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is comprehensive but overly verbose for its purpose — it reads more like a documentation page than a concise skill instruction. It contains substantial informational content (model behavior differences, enterprise considerations, cost optimization) that Claude doesn't need to effectively help users switch or configure models. The actionable core (how to switch models, BYOK setup) is buried among advisory content.
Suggestions
Cut the content by ~60%: remove the code generation style comparison, enterprise considerations, and detailed cost optimization sections — these are informational, not actionable for the task of configuring models.
Add a verification step after BYOK configuration (e.g., 'Send a test message in Chat and confirm the model name appears in the response header').
Move the model tables and task-selection quick reference to the top as the primary content, since that's the most actionable part for model selection.
Remove or drastically shorten the 'Model Behavior Differences' section — Claude already knows how different models behave, and this information changes frequently.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Significant verbosity throughout. The code generation style comparison section explains things Claude already knows about its own behavior vs GPT models. The enterprise considerations, cost optimization strategies, and quota management details are padding that Claude doesn't need to perform model selection. Model tables include information that changes frequently and could be outdated. | 1 / 3 |
Actionability | The 'How to Switch Models' section provides concrete UI navigation steps, and the BYOK configuration shows specific key formats and settings paths. However, much of the content is informational/advisory rather than executable — the task selection guide is a recommendation table, not actionable instructions, and there are no commands or scripts to run. | 2 / 3 |
Workflow Clarity | The quick reference for model selection by task is clear, and the BYOK setup has numbered steps. However, there's no validation checkpoint — e.g., how to verify a BYOK key is working, how to confirm the model actually switched, or what to do if a model request fails. For a configuration task, verification steps are important. | 2 / 3 |
Progressive Disclosure | The content is organized with clear headers and sections, and external resource links are provided at the end. However, this is a long monolithic file (~150+ lines of content) with sections like enterprise considerations, code generation style comparisons, and cost optimization that could be split into separate reference files. No bundle files exist to offload this content. | 2 / 3 |
Total | 7 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
allowed_tools_field | 'allowed-tools' contains unusual tool name(s) | Warning |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 9 / 11 Passed | |
- Commit
c81cea4
- 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.