skill-creator

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

Quality

71%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/public/skill-creator/SKILL.md

Quality

Content

56%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.

The body is highly actionable with clear, validated workflows, but it is padded with conversational filler and repeated summaries that hurt token efficiency. Navigation is undermined by broken references and un-signaled bundle files, and it runs past its own recommended length.

Suggestions

Cut the chatty asides and the duplicate core-loop summary at the end to bring the body under 500 lines and improve conciseness.

Fix broken references — create the referenced `agents/grader.md`, `agents/analyzer.md`, `agents/comparator.md` and `eval-viewer/generate_review.py`, or remove/redirect the citations to actual files like `scripts/generate_report.py`.

Add explicit pointers in the body to the existing but currently un-referenced bundle files (`references/output-patterns.md`, `references/workflows.md`, and the individual scripts) so each resource is clearly signaled and discoverable.

Dimension	Reasoning	Score
Conciseness	At ~535 lines the body is noticeably verbose, padded with chatty asides ('Cool? Cool.', 'who knows', the plumbers/grandparents anecdote, 'billions a year in economic value') and restates the core loop twice (intro and closing), adding little Claude-doesn't-know information.	2 / 5
Actionability	Provides concrete executable guidance throughout — exact `python -m scripts.*` commands, JSON schemas, file paths, and viewer launch flags — with only minor gaps where referenced scripts are not present in the bundle.	4 / 5
Workflow Clarity	Multi-step processes (Step 1–5 run/eval, iteration loop, description optimization) are clearly sequenced with checkpoints and feedback loops; minor gaps include relying on a non-existent `agents/grader.md` for grading.	4 / 5
Progressive Disclosure	Structure is present but flawed: the body references non-existent paths (`agents/*.md`, `eval-viewer/generate_review.py`) while real bundle files (`references/output-patterns.md`, `references/workflows.md`, several scripts) go unmentioned, and at 535 lines it exceeds its own stated 500-line ideal, signaling content that should be split out.	3 / 5
	Total	13 / 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.

The description is strong: it clearly communicates both capability and triggering context using concrete, natural-language terms with an explicit 'Use when...' clause. Minor tightening could replace the slightly generic 'measure skill performance' phrasing.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ('Create new skills, modify and improve existing skills', 'run evals to test a skill', 'benchmark skill performance with variance analysis', 'optimize a skill's description'), with only minor generic gaps such as 'measure skill performance'.	4 / 5
Completeness	Explicitly states both what it does (create/modify/measure skills) and when to use it via a concrete 'Use when users want to...' clause enumerating triggering scenarios.	5 / 5
Trigger Term Quality	Includes natural user phrases ('create a skill from scratch', 'edit', 'optimize an existing skill', 'run evals', 'benchmark') with good coverage, though a few common synonyms are absent.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear meta-tooling niche (skill authoring/optimization) with distinct trigger phrases that are unlikely to collide with ordinary coding or document skills.	5 / 5
	Total	18 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (535 lines); consider splitting into references/ and linking	Warning
referenced_paths_exist	Referenced path issues: 2 missing	Warning

	Total	14 / 16 Passed

Repository: bytedance/deer-flow
Path: skills/public/skill-creator/SKILL.md
Commit: 2a261d2

Reviewed: about 18 hours 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.