skill-creator

Create, revise, evaluate, publish, and improve Open-Science Skills through the native JavaScript host.skills composer. Use when the user wants a reusable workflow, an existing Skill changed, test cases or benchmarks for a Skill, or better Skill triggering.

Quality

88%

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

85%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.

An unusually disciplined body: lean, concrete, and well-sequenced with real validation checkpoints around destructive operations. Its main defect is bundle hygiene — four referenced paths point to files that do not exist, and about half the shipped scripts and the asset are orphaned with no navigation from SKILL.md.

Suggestions

Fix broken references: add the missing `agents/grader.md`, `agents/comparator.md`, `agents/analyzer.md`, and `eval-viewer/generate-review.js` files, or rewire steps 5-9 of the evaluate section to the files that actually exist (e.g. `scripts/generate-report.js`).

Add a short bundle map (or inline links) for the orphaned files — `scripts/quick-validate.js`, `scripts/run-eval.js`, `scripts/run-loop.js`, and `assets/eval_review.html` — so an agent can discover them from SKILL.md.

Replace the `host.agents.attachSkill(...)` placeholder with the actual call signature (or an explicit note on where it is documented) so the attach workflow is copy-paste executable.

Dimension	Reasoning	Score
Conciseness	The body is lean and imperative with no explanations of concepts Claude already knows; every section (composer semantics, stage selection, authoring rules, eval gating, improvement heuristics) carries operational content that earns its tokens.	5 / 5
Actionability	Provides concrete executable guidance (the full `host.skills` API list, exact paths like `evals/evals.json` and `trigger-evals.json`, named scripts to run), but four referenced files (`agents/grader.md`, `agents/comparator.md`, `agents/analyzer.md`, `eval-viewer/generate-review.js`) do not exist in the bundle and `host.agents.attachSkill(...)` is a placeholder, fitting the minor-gaps anchor.	4 / 5
Workflow Clarity	Six stages are clearly sequenced with explicit validation checkpoints ("Re-read changed files, call `host.skills.validate(name)`", "Publish only after the user accepts the draft", "Read the published `SKILL.md` back") and guarded destructive operations (exact `draft-<name>`/`personal-<name>` IDs, `overwrite = true` only on explicit user choice), so the destructive-operation cap does not apply.	5 / 5
Progressive Disclosure	In-file structure is good with one-level links, but scored against the actual bundle: `agents/grader.md`, `agents/comparator.md`, `agents/analyzer.md`, and `eval-viewer/generate-review.js` are referenced yet missing, while `quick-validate.js`, `run-eval.js`, `run-loop.js`, `generate-report.js`, and `assets/eval_review.html` exist but are never referenced — broken navigation and undiscoverable files exceed the "minor organization gaps" of the anchor at 4.	3 / 5
	Total	17 / 20 Passed

Description

88%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 that states concrete capabilities and an explicit, well-phrased trigger clause. Its only weaknesses are a slightly generic "reusable workflow" trigger and missing common synonyms like "write/edit a skill" that would sharpen matching and reduce overlap risk.

Dimension	Reasoning	Score
Specificity	Lists five concrete lifecycle actions ("Create, revise, evaluate, publish, and improve") plus a concrete mechanism ("through the native JavaScript host.skills composer"), matching the comprehensive-coverage anchor rather than the minor-gaps anchor at 4.	5 / 5
Completeness	Explicitly answers both what ("Create, revise, evaluate, publish, and improve ... Skills through the native JavaScript host.skills composer") and when ("Use when the user wants a reusable workflow, an existing Skill changed, test cases or benchmarks for a Skill, or better Skill triggering") with concrete trigger phrases.	5 / 5
Trigger Term Quality	Includes natural phrases users would say ("a reusable workflow", "an existing Skill changed", "test cases or benchmarks for a Skill", "better Skill triggering") but omits common variations such as "write/edit/update a skill", "SKILL.md", or "skill won't fire", fitting the good-coverage-with-a-few-missing anchor rather than comprehensive synonym coverage.	4 / 5
Distinctiveness Conflict Risk	The skill-lifecycle niche is mostly distinct with skill-specific triggers, but "a reusable workflow" is generic enough to risk overlap with unrelated workflow requests (e.g. CI or general automation), matching the minor-overlap anchor.	4 / 5
	Total	18 / 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
relative_links	Relative link issues: 3 missing	Warning

	Total	15 / 16 Passed

Repository: aipoch/open-science
Path: resources/skills/skill-creator/SKILL.md
Commit: bf35648

Reviewed: 8 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.