CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

doc-gardening

Find docs and agent instructions in the carnival that no longer match the code, and file one GitHub issue per stale doc. Use when running the doc gardening maintenance loop, or when asked to check whether the README, AGENTS.md, a game's README or a skill is out of date. Files issues only; never changes code or docs.

Quality

95%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

93%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 exemplary instruction-only skill body: concise, immediately actionable with concrete templates and commands, and well-organized for a single-file skill. The single gap is workflow hardening — no explicit step ordering or post-filing verification loop for the batch issue-filing operation.

DimensionReasoningScore

Conciseness

Lean throughout: a two-sentence motivation ('Docs drift when code changes... a stale one leads them astray'), then only operational content — safety boundaries, what to check, what to file. No explanation of concepts Claude already knows and no padding, matching the level-5 anchor where every token earns its place.

5 / 5

Actionability

Gives copy-paste-ready guidance: the issue title template 'Docs: <file> is out of date', a body format with a concrete line-pointer example 'src/games/water-cannon/game.ts:12', the verification command 'npm run check', and exact paths ('src/games/<game-id>/README.md', 'skills/', 'tessl-factory/skills/'). For an instruction-only skill this fully matches the level-5 anchor.

5 / 5

Workflow Clarity

The sequence is clear (read each doc, check claims against code, dedup against open issues, file one issue per stale doc) and key checkpoints are explicit ('Before filing, check the open issues and skip', 'Run commands like `npm run check` if you need to confirm one works', 'Nothing found is a fine result'). Held at 4 rather than 5 because steps are conveyed through section order rather than an explicit sequence, and there is no post-filing verification or feedback loop for this batch write operation.

4 / 5

Progressive Disclosure

A single self-contained file (~46 lines) with no bundle directories and well-organized sections ('Staying safe', 'What to check', 'What to file'). Per the rubric's guideline for skills under 50 lines with no need for external references, this scores 5 on organization alone.

5 / 5

Total

19

/

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.

A well-crafted description that states concrete capabilities, includes an explicit 'Use when' clause with natural trigger phrases and specific file names, and carves out a distinct niche with a clear write-scope boundary. The only weakness is modest action coverage beyond the two core verbs (find and file).

DimensionReasoningScore

Specificity

Names the domain and several concrete actions — 'Find docs and agent instructions... that no longer match the code, and file one GitHub issue per stale doc' — plus enumerated targets (README, AGENTS.md, game READMEs, skills) and a write boundary ('Files issues only; never changes code or docs'). Scored 4 rather than 5 because the action list stops at two core verbs, leaving minor coverage gaps (e.g., how claims get verified), and clearly exceeds the 1-2-action level 3 anchor.

4 / 5

Completeness

Explicitly answers both questions: the 'what' ('Find docs... that no longer match the code, and file one GitHub issue per stale doc') and the 'when' ('Use when running the doc gardening maintenance loop, or when asked to check whether the README, AGENTS.md, a game's README or a skill is out of date'). The 'Use when' clause with concrete trigger phrases matches the level-5 anchor exactly.

5 / 5

Trigger Term Quality

Covers the natural phrases a user would actually say — 'out of date', 'stale doc', 'doc gardening maintenance loop' — alongside specific file names ('README', 'AGENTS.md', 'a game's README', 'a skill'), giving comprehensive coverage including synonyms. This matches the level-5 anchor better than level 4 because both synonyms (stale / out of date) and concrete file names are present.

5 / 5

Distinctiveness Conflict Risk

A clear niche — code/doc staleness checking plus issue filing in a specific repo ('the carnival') — with the explicit boundary 'Files issues only; never changes code or docs' distinguishing it from doc-editing or general review skills. Minimal conflict risk, matching the level-5 anchor rather than the minor-overlap level-4 anchor.

5 / 5

Total

19

/

20

Passed

Validation

100%

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

Cornebidouille/tessl-carnival

Path

tessl-factory/skills/doc-gardening/SKILL.md

Reviewed

Table of Contents

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.