CtrlK
BlogDocsLog inGet started
Tessl Logo

self-customize

Customize your own agent — add capabilities, install packages, add MCP servers, edit code or CLAUDE.md. Use when the user asks you to add a feature, install a tool, or modify how you work. For non-trivial code changes, delegate to a builder agent via create_agent.

70

Quality

85%

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

82%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, highly actionable skill body with concrete tool calls and clear workflows. The main improvement opportunities are trimming the rationale section and adding an explicit validate-fix-retry loop for failed operations.

Suggestions

Tighten or relocate the "Diff Size Limits — Why" section — the limits themselves are actionable, but the rationale largely explains what Claude can infer.

Add an explicit validate-then-fix retry loop for when install_packages approval is denied or a builder reports a failure, to push workflow_clarity to 5.

Consider moving the reusable builder-instructions block into a reference file (e.g. references/builder-prompt.md) so the main body stays a lean overview.

DimensionReasoningScore

Conciseness

Mostly lean and assumes Claude's competence (references bun, container restarts, composed provider docs without explaining them), but the "Diff Size Limits — Why" section is rationale Claude could largely infer, a minor instance of over-explanation that could be trimmed.

4 / 5

Actionability

Fully executable guidance with concrete tool calls and arguments (e.g. install_packages({ apt: ["ffmpeg"], npm: ["@xenova/transformers"], reason: "..." }), add_mcp_server({ name: "rss", command: "npx", args: [...] })), a copy-paste-ready builder instructions block, and worked examples covering the common cases.

5 / 5

Workflow Clarity

Clear sequenced workflows (decision tree, 5-step builder flow, numbered examples) with checkpoints (admin approval, builder report review, "Test the new capability once the container restarts"); not a 5 because there is no explicit validate-then-fix retry loop for failed installs or builder errors.

4 / 5

Progressive Disclosure

Well-organized into clear sections (Decision Tree, Workflow, Builder Instructions, Why, Examples, When NOT to) with no external references needed; the inline builder-instructions block could arguably live in a separate reusable reference file, a minor organization gap keeping it below 5.

4 / 5

Total

17

/

20

Passed

Description

87%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 explicit what/when structure and concrete, distinct triggers. Its only real weakness is the second-person framing ("Customize your own agent") rather than third-person voice.

DimensionReasoningScore

Specificity

Lists several concrete actions ("add capabilities, install packages, add MCP servers, edit code or CLAUDE.md", "delegate to a builder agent via create_agent") but uses second-person voice ("Customize your own agent"), which per the guidelines reduces the specificity score by one from 5.

4 / 5

Completeness

Explicitly answers both what (capabilities/packages/MCP/code/CLAUDE.md changes, builder delegation) and when ("Use when the user asks you to add a feature, install a tool, or modify how you work") with concrete trigger phrases.

5 / 5

Trigger Term Quality

Natural user phrases like "add a feature", "install a tool", and "modify how you work" are present and likely to be said, but coverage is somewhat generic and lacks synonyms or specific file extensions, so a few natural terms are missing.

4 / 5

Distinctiveness Conflict Risk

Occupies a clear niche (self-modifying the agent's own environment) with distinct triggers (adding features, installing tools, MCP servers, editing code/CLAUDE.md) and minimal overlap risk with other skills.

5 / 5

Total

18

/

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.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
nanocoai/nanoclaw
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.