refine

End-of-session reflection. Reviews friction encountered during the session and proposes updates to docs/ to capture lessons learned.

Quality

75%

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 ./.agents/skills/refine/SKILL.md

Quality

Content

86%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.

This is a well-crafted, instruction-only skill body: lean, concrete, and clearly sequenced, with rules that enforce session-verified knowledge and minimal edits. The only meaningful gap is the absence of an explicit verification checkpoint and a worked example of a doc update.

Dimension	Reasoning	Score
Conciseness	The 38-line body is lean and assumes Claude's competence: no concept explanations, each instruction adds operational value, and the Rules section actively forbids padding ('add a line or two, not a paragraph', 'Don't add things Claude already knows'). This matches the 'every token earns its place' anchor.	5 / 5
Actionability	Concrete, executable guidance throughout: a copy-paste command (`git diff main...HEAD --stat`), named files to read (docs/validation.md, docs/enterprise.md), and a categorized decision rubric for updates. It falls short of 5 only because no example of an actual before/after doc update is shown, leaving the 'apply updates' step's output format implicit.	4 / 5
Workflow Clarity	A clear five-step sequence (identify friction → read docs → propose updates → apply → report) with well-defined steps and scope-constraining rules ('only add knowledge confirmed by this session'). Doc edits are low-risk so the destructive-operation cap does not apply, but there is no explicit verification checkpoint (e.g., re-reading the edited doc or confirming consistency), which keeps it below 5.	4 / 5
Progressive Disclosure	Per the simple-skill guideline, a sub-50-line skill with no need for external references scores 5 with well-organized sections: the body has clean Instructions and Rules sections, no inlined content that belongs in separate files, and no nested references.	5 / 5
	Total	18 / 20 Passed

Description

53%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 concise and in third person with a clear 'what', but it lacks an explicit 'Use when...' trigger clause and its action coverage and keyword set are moderate. Adding explicit trigger guidance and a few natural user phrases would lift it substantially.

Suggestions

Add an explicit trigger clause, e.g., 'Use when the user asks to reflect on the session, capture lessons learned, or update docs after finishing work.'

Replace the abstract noun 'friction' with concrete failure language users would actually say (e.g., 'what was slow, what failed, what took multiple attempts').

List the concrete update targets by name (validation.md, enterprise.md, CLAUDE.md pointers) to sharpen the 'what' and improve trigger-term coverage.

Dimension	Reasoning	Score
Specificity	Names the domain ('End-of-session reflection') and two concrete actions ('Reviews friction encountered during the session' and 'proposes updates to docs/'), but 'friction' is abstract and the action list is not comprehensive, matching the '1-2 concrete actions' anchor rather than the 'several specific actions' anchor above.	3 / 5
Completeness	The 'what' is clear (reviews session friction, proposes docs/ updates), but there is no explicit 'Use when...' clause or equivalent trigger guidance — 'End-of-session' only weakly implies timing. Per the judging guidelines, a missing explicit trigger clause caps completeness at 3.	3 / 5
Trigger Term Quality	Some relevant keywords are present ('End-of-session', 'reflection', 'friction', 'lessons learned'), but common variations users would naturally say (e.g., 'retro', 'what went wrong', 'update the docs', 'session summary') are missing. It sits between the 'some relevant keywords' (3) and 'good keyword coverage' (4) anchors but is noticeably closer to 3.	3 / 5
Distinctiveness Conflict Risk	The end-of-session reflection + docs-update niche is distinct from typical build/analyze skills and unlikely to trigger wrongly. Minor overlap risk with general documentation or code-review skills keeps it at 4 rather than the 'clear niche, minimal conflict' 5 anchor.	4 / 5
	Total	13 / 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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: windmill-labs/windmill
Path: .agents/skills/refine/SKILL.md
Commit: af42c47

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.