CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

tooluniverse-install-skills

Detect and auto-install missing ToolUniverse research skills. Checks common Claude Code/Cursor/Codex skill directories for the canary file, and installs any missing skills if none found. Use when the plugin's research skills aren't loading, when migrating between clients, or when verifying a skill installation.

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

78%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 tight, executable skill: the detection and installation commands are copy-paste ready and the workflow has an explicit decision branch plus a confirmation step. The main gaps are the Cursor-hardcoded success check, no client-detection command despite a detection table, and no error-recovery guidance for failed installs.

Suggestions

Make the post-install confirmation client-generic (e.g., re-run the detection command from the first section and assert the output is a file path, rather than the hardcoded '.cursor/...' ls).

Add a brief failure path for Installation: what to do if the clone/checkout fails or the canary file is still missing after copying (e.g., report the exact directory used and suggest re-running detection).

Provide a one-line command for client detection (e.g., ls -d .cursor .claude .agents ... 2>/dev/null) so the Client Detection table is actionable rather than only descriptive, and drop the duplicated per-client comments in the Installation block to remove redundancy.

DimensionReasoningScore

Conciseness

The body is lean — copy-paste bash with terse comments and no explanations of concepts Claude already knows. Not a 5: the client→directory mapping appears twice (per-client comments in the Installation block and again in the Client Detection table), and the opening line restates the frontmatter description.

4 / 5

Actionability

The detection command and sparse-checkout install block are fully executable and copy-paste ready. Not a 5: the success-confirmation step hardcodes the Cursor path ("ls .cursor/skills/tooluniverse-drug-research/SKILL.md") and doesn't generalize to the other seven clients, and there's no command to detect which client config directory exists.

4 / 5

Workflow Clarity

Clear sequence — detect (with an explicit branch: file path → stop; NOT_INSTALLED → install), install, then confirm — with a post-install verification checkpoint, so the destructive/batch cap does not apply. Not a 5: no error-recovery loop for likely failures (clone fails, canary still missing after install, ambiguous client when multiple config dirs exist).

4 / 5

Progressive Disclosure

A single-page, single-purpose skill (~70 lines, mostly code) that needs no external references, with four well-organized sections (Detection, Installation, Client Detection, After Installation). Per the rubric's simple-skill guidance, well-organized sections with no reference indirection earns the top anchor.

5 / 5

Total

17

/

20

Passed

Description

87%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 strong description: concrete actions, explicit 'Use when...' triggers covering load failure, client migration, and installation verification, and a clearly scoped niche. Only minor gaps — a few natural synonym triggers are absent and the actions are summarized at a high level.

DimensionReasoningScore

Specificity

"Detect and auto-install missing ToolUniverse research skills. Checks common Claude Code/Cursor/Codex skill directories for the canary file, and installs any missing skills if none found" lists several concrete actions (detect, check directories for canary file, install). Not a 5: coverage has minor gaps — it never says what the skills cover or how installation is performed, and 'auto-install' summarizes rather than names the action.

4 / 5

Completeness

Explicitly answers both: what ("Detect and auto-install missing ToolUniverse research skills... checks common... skill directories for the canary file") and when ("Use when the plugin's research skills aren't loading, when migrating between clients, or when verifying a skill installation") with three concrete trigger phrases. Clearly matches the top anchor, not the 4 anchor where 'when' would be less explicit.

5 / 5

Trigger Term Quality

Triggers like "when the plugin's research skills aren't loading", "when migrating between clients", and "verifying a skill installation" are natural phrases users would say. Not a 5: common variations like 'skills not found', 'install ToolUniverse', or directory keywords are missing.

4 / 5

Distinctiveness Conflict Risk

A clear niche — installing one named plugin's (ToolUniverse) research skills — with distinct triggers tied to that plugin. Overlap with generic skill-install requests is minimal because the description scopes itself to the ToolUniverse plugin and its canary file.

5 / 5

Total

18

/

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

mims-harvard/ToolUniverse

Path

plugin/skills/tooluniverse-install-skills/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.