CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

agents-md

This skill should be used when the user asks to "create AGENTS.md", "update AGENTS.md", "maintain agent docs", "set up CLAUDE.md", or needs to keep agent instructions concise. Enforces research-backed best practices for minimal, high-signal agent documentation.

Quality

80%

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

Fix and improve this skill with Tessl

tessl review fix ./plugins/AI-Agents-Safe-Coding-Skills-claude/skills/agents-md/SKILL.md

The canonical home for this skill is agents-md in sickn33/agentic-awesome-skills

SKILL.md
Quality
Evals
Security

Quality

Content

75%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.

A well-structured, actionable, token-efficient instruction skill with a clear create-analyze-write workflow and concrete templates throughout. The main defects are the broken nested code fencing in the Commit Attribution example, a ~25-line duplicated Example Structure, and the absence of any validation checkpoint for the workflow.

Suggestions

Fix the Commit Attribution and Example Structure blocks: the inner triple-backtick fences close the outer markdown block early. Use a four-backtick outer fence (or indent the inner block) so the templates render and copy correctly.

Add an explicit validation step after writing, e.g., 'Count lines: `wc -l AGENTS.md`. If over 60, cut; never ship over 100' — this both enforces the stated budget and supplies the missing workflow checkpoint.

Remove the duplicated Example Structure or reduce it to a bare section-order outline ('Package Manager → Commit Attribution → File-Scoped Commands → API Routes → CLI'), since every template it contains already appears in Required Sections.

DimensionReasoningScore

Conciseness

The body is lean: headers plus bullets, concrete file names to check, no filler or pleasantries, and no explanation of concepts Claude already knows. What keeps it below a 5 is the ~25-line Example Structure that duplicates the Required Sections templates (Commit Attribution and the File-Scoped Commands table appear twice verbatim), plus scattered one-line rationales ('Per-file commands are faster and cheaper than full project builds') that could be trimmed.

4 / 5

Actionability

Guidance is mostly executable and copy-paste ready: the symlink command `ln -s AGENTS.md CLAUDE.md`, specific lock-file names, and concrete markdown templates/tables for each section. The gap is the Commit Attribution template: its nested triple-backtick fences are unescaped, so the example block terminates early and renders broken — that block is not clean copy-paste material, holding it below the fully-executable anchor 5.

4 / 5

Workflow Clarity

The sequence is clear and ordered: File Setup (create file, symlink) → Before Writing (project analysis with concrete checks) → Writing Rules → Required/Optional sections → Anti-Patterns → Example. However, there is no validation checkpoint anywhere (e.g., verify line count is under 60/100 after writing, or confirm the symlink resolved), so it sits at 'clear sequence with most checkpoints present; minor validation gaps' rather than the anchor-5 explicit validation step.

4 / 5

Progressive Disclosure

No bundle files exist and none are needed — this is an instruction-only skill whose content fits appropriately in one file, with well-labeled sections and clear internal organization. It exceeds the under-50-lines threshold for an automatic 5, and a reviewer could argue the full Example Structure belongs in a separate reference file, so it lands at 'good structure; most content appropriately placed; minor organization gaps'.

4 / 5

Total

16

/

20

Passed

Description

77%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 strong description with excellent, explicitly quoted trigger phrases and explicit use-when guidance. Its main weakness is the capability statement, which is one abstract clause ('enforces research-backed best practices') rather than concrete operations.

Suggestions

Replace the abstract capability clause with concrete actions, e.g., 'Creates and maintains a minimal AGENTS.md at project root (with a CLAUDE.md symlink) containing package manager, file-scoped commands, commit attribution, and key conventions.'

Drop the buzzword hedge 'research-backed' and state the enforceable constraint directly ('caps the file at 60 lines, 100 hard maximum').

DimensionReasoningScore

Specificity

The capability statement is a single abstract clause — 'Enforces research-backed best practices for minimal, high-signal agent documentation' — which names the domain but leans on buzzword-ish language ('research-backed', 'high-signal'). Concrete actions (create, update, maintain, keep concise) only surface through the user-trigger verbs, matching the anchor 'names domain and 1-2 concrete actions, but not comprehensive'. A 4 would require the skill itself listing several specific operations it performs.

3 / 5

Completeness

Both parts are present: the 'when' is explicit and concrete ('This skill should be used when the user asks to...'), and the 'what' is stated ('Enforces research-backed best practices for minimal, high-signal agent documentation'). However, the 'what' is vague about what the skill actually produces, falling short of the anchor-5 requirement that both halves be clearly and explicitly concrete.

4 / 5

Trigger Term Quality

Quoted natural phrases users would actually say — 'create AGENTS.md', 'update AGENTS.md', 'maintain agent docs', 'set up CLAUDE.md' — plus synonyms (agent docs, agent instructions) and both file names (AGENTS.md, CLAUDE.md). This matches the anchor for comprehensive natural-term coverage including synonyms.

5 / 5

Distinctiveness Conflict Risk

AGENTS.md/CLAUDE.md maintenance is a clear, distinct niche with dedicated trigger phrases. Minor overlap risk with generic CLAUDE.md-initialization or documentation skills via 'set up CLAUDE.md', matching 'mostly distinct; minor overlap risk with closely related skills'.

4 / 5

Total

16

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

administrakt0r/AI-Agents-Safe-Coding-Skills

Path

plugins/AI-Agents-Safe-Coding-Skills-claude/skills/agents-md/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.