CtrlK
BlogDocsLog inGet started
Tessl Logo

shell-builtins

Use when creating a new shell builtin command for Crush (internal/shell/), editing an existing one, or when the user needs to understand how commands are intercepted in Crush's embedded shell.

70

Quality

88%

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 lean skill body: dense with non-obvious project knowledge, fully actionable code templates, and a clear step sequence. The only gap is the absence of a verification step for confirming a newly added builtin works.

DimensionReasoningScore

Conciseness

Every sentence carries non-obvious, project-specific facts — e.g. "registered in standardHandlers() before the block handler", "the stateless Run entrypoint used by the hook runner", and the ctx.Err() vs interp.ExitStatus(n) semantics. No generic shell/Go padding; anchor 5 (every token earns its place).

5 / 5

Actionability

Each step ships copy-paste-adaptable code: the `case "mycommand":` snippet, the exact handler signature `func handleMyCommand(ctx context.Context, args []string, stdin io.Reader, stdout, stderr io.Writer) error`, and the ctx.Err() polling loop. The common case (adding a builtin) is fully covered with specific file paths.

5 / 5

Workflow Clarity

A clear six-step numbered sequence with rationale at each step, but no explicit verification checkpoint (nothing about how to test that the new builtin works). Fits anchor 4 (clear sequence, minor validation gaps); the destructive/batch cap does not apply since this is additive code with compiler feedback.

4 / 5

Progressive Disclosure

The skill is self-contained with no bundle files and no external references, organized into clean sections (How Builtins Work, Adding a New Builtin, Existing Builtins table). Per the rubric's simple-skill guidance, well-organized sections with no nested references score 5.

5 / 5

Total

19

/

20

Passed

Description

78%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, niche-specific description with explicit triggers and concrete named actions. Its main weakness is that the 'what' is only implied by the trigger clauses rather than stated up front, and a few natural synonyms are missing.

Suggestions

Lead with a short third-person what-statement (e.g., "Guides adding and editing in-process builtin commands in Crush's embedded shell") before the "Use when..." clause so both elements are explicitly stated.

Add one or two natural trigger variations users might say, such as "add a command to Crush's shell" or "bash tool builtins", to broaden keyword coverage.

Optionally name the intercept mechanism in the description ("commands intercepted before reaching the OS") to sharpen the what-portion.

DimensionReasoningScore

Specificity

Names three concrete actions ("creating a new shell builtin command", "editing an existing one", "understand how commands are intercepted") plus concrete paths like internal/shell/, matching the 'several specific actions with minor gaps' anchor. Not 5 because it never states what the skill itself provides (handler wiring, code patterns).

4 / 5

Completeness

An explicit "Use when..." clause with three distinct triggers is present, and the 'what' is embedded in the gerund phrases rather than stated as a standalone capability sentence. Fits anchor 4 (both present, one could be more explicit — here the what); not 5 since there is no leading what-statement as in the anchor-5 example.

4 / 5

Trigger Term Quality

Good natural keyword coverage — "shell builtin", "command", "Crush", "intercepted", "embedded shell" — all terms a user would plausibly say. Not 5 because common phrasings like "add a command to the shell" or "bash tool" are absent.

4 / 5

Distinctiveness Conflict Risk

"Crush (internal/shell/)", "shell builtin", and "embedded shell" carve out a clear niche with distinct triggers and minimal overlap with generic shell or command skills. Matches anchor 5.

5 / 5

Total

17

/

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
charmbracelet/crush
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.