skill-creator

Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Gemini CLI's capabilities with specialized knowledge, workflows, or tool integrations.

Quality

66%

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 ./packages/core/src/skills/builtin/skill-creator/SKILL.md

Quality

Content

63%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 delivers a clear, actionable seven-step skill-creation workflow with real CLI commands and a validation feedback loop. Its main weaknesses are conciseness padding, dangling references to missing reference files, and a referenced "available_resources" section that is absent.

Suggestions

Remove or relocate the conceptual padding (conciseness preaching, the bridge/field analogy, "What Skills Provide") to trim the body well under 500 lines, per the skill's own guidance.

Add the referenced references/workflows.md and references/output-patterns.md files (or remove the pointers to them), since the references/ directory is currently missing and the pointers dangle.

Define the "available_resources" section referenced by Steps 3 and 5 (with absolute script paths), or rewrite those commands so they are copy-paste ready without it.

Dimension	Reasoning	Score
Conciseness	Substantive procedural content (7-step process, exact CLI commands, naming rules) is offset by noticeable padding: repeated conciseness preaching, the "narrow bridge with cliffs" analogy, and conceptual exposition of what skills/scripts/references are that the model already knows; the body sits at its own 500-line ceiling.	3 / 5
Actionability	Provides concrete executable commands (node init_skill.cjs, node package_skill.cjs, gemini skills install --scope) and specific naming rules; falls short of fully copy-paste ready because commands use placeholders (<path-to-skill-creator>) and reference an "available_resources" section that does not appear in the body.	4 / 5
Workflow Clarity	A clearly sequenced 7-step workflow with an explicit validation feedback loop at packaging ("If validation fails... Fix any validation errors and run the packaging command again") and a /skills list verification step; not a 5 because the validation is delegated to the script and described in prose rather than structured as explicit gated checkpoints in the step flow.	4 / 5
Progressive Disclosure	Sections are well-organized and the referenced scripts (init/package/validate_skill.cjs) exist in the bundle, but the body directs readers to references/workflows.md and references/output-patterns.md that do not exist (the references/ directory is missing), and conceptual material is inlined at the 500-line limit.	3 / 5
	Total	14 / 20 Passed

Description

70%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 clearly states the skill's purpose and provides an explicit, natural-language trigger for when to use it, with low conflict risk. Its main weakness is a generic "what" that under-enumerates the concrete authoring actions the skill actually performs.

Suggestions

Expand the "what" to list concrete actions the skill performs (e.g. scaffold, write SKILL.md, package, validate, install) rather than the generic "Guide for creating effective skills".

Add a couple of natural trigger synonyms (e.g. "build a skill", "make a skill for Gemini CLI") to broaden keyword coverage.

Tighten or drop the broad "extends Gemini CLI's capabilities with specialized knowledge, workflows, or tool integrations" clause, which describes what skills are rather than what this skill does and adds overlap surface.

Dimension	Reasoning	Score
Specificity	Names the skill-authoring domain and two concrete actions ("create a new skill (or update an existing skill)"), but the "what" ("Guide for creating effective skills") is generic and does not enumerate several specific actions like design, scaffold, validate, or package.	3 / 5
Completeness	Explicitly answers both what ("Guide for creating effective skills") and when ("This skill should be used when users want to create a new skill (or update an existing skill)"), satisfying the trigger-clause requirement; not a 5 because the "what" lacks the concrete multi-action enumeration the top anchor requires.	4 / 5
Trigger Term Quality	Includes natural phrases a user would say ("create a new skill", "update an existing skill", "skill", "workflows", "tool integrations") with good coverage; a few natural variants (e.g. "build a skill", "make a skill") are missing, so it stops short of comprehensive.	4 / 5
Distinctiveness Conflict Risk	Skill-authoring is a distinct niche with low conflict risk, but the broad phrasing "extends Gemini CLI's capabilities with specialized knowledge, workflows, or tool integrations" adds some overlap surface with general skill/plugin tooling.	4 / 5
	Total	15 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (520 lines); consider splitting into references/ and linking	Warning
relative_links	Relative link issues: 6 missing	Warning
referenced_paths_exist	Referenced path issues: 17 missing	Warning

	Total	13 / 16 Passed

Repository: google-gemini/gemini-cli
Path: packages/core/src/skills/builtin/skill-creator/SKILL.md
Commit: 9a15c45

Reviewed: 1 day 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.