research

Research a documentation topic — locate affected files, understand the problem, identify what to change. Use when investigating an issue, a question, or a topic before writing a fix. Triggers on: "research issue 1234", "investigate what needs changing for #500", "what files are affected by #200", "where is X documented", "is our docs page about Y accurate", "look into how we document Z".

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

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.

The body is a concise, well-sequenced research workflow with strong progressive disclosure and a real verification checkpoint. Its main weakness is actionability: beyond the single gh command, steps are instructional rather than executable.

Suggestions

Add concrete search examples for step 2 (e.g., a ripgrep/grep command against content/ with the /manuals prefix mapping) so locating affected files is executable rather than descriptive.

Give a concrete fetch command or snippet for step 6's live-site check instead of a bare URL block.

Provide a minimal findings-report template or required fields list in step 7 to make the output format actionable for the write step.

Dimension	Reasoning	Score
Conciseness	Lean and efficient; assumes Claude's competence with no concept over-explanation, and every section (Gather context through Report findings) earns its place.	5 / 5
Actionability	Only one concrete executable command is given (gh issue view); most steps rely on vague direction ("Search content/", "Follow external links") without specific commands or examples, matching the some-concrete-guidance-but-incomplete anchor.	3 / 5
Workflow Clarity	Clear numbered sequence with an explicit verification checkpoint in section 5 ("Do not plan a fix based on an unverified claim") and blocker reporting, but the verify-fail-retry feedback loop is lighter than the validate-fix-revalidate anchor of 5.	4 / 5
Progressive Disclosure	Cleanly organized into well-scoped numbered sections with a one-level-deep external reference (the CLAUDE.md vendored content table); no bundle files exist and none are needed for this single-purpose research skill.	5 / 5
	Total	17 / 20 Passed

Description

95%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 well-crafted: it states concrete actions, gives both the what and when, and lists realistic natural-language triggers that make it highly distinct. Minor gain available by making the action verbs more granular.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ("locate affected files, understand the problem, identify what to change") with minor abstraction in "understand the problem", falling just below the comprehensive multi-action anchor of 5.	4 / 5
Completeness	Explicitly answers both what ("locate affected files, understand the problem, identify what to change") and when ("Use when investigating an issue ... before writing a fix") with concrete trigger phrases.	5 / 5
Trigger Term Quality	Provides many natural phrases a user would actually say ("research issue 1234", "where is X documented", "is our docs page about Y accurate"), matching the comprehensive-coverage anchor.	5 / 5
Distinctiveness Conflict Risk	Clear niche (docs research prior to a fix) with distinct, specific triggers that are unlikely to fire for unrelated skills, matching the minimal-conflict-risk anchor.	5 / 5
	Total	19 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: docker/docs
Path: .agents/skills/research/SKILL.md
Commit: ff96ad1

Reviewed: 3 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.