CtrlK
BlogDocsLog inGet started
Tessl Logo

command-development

This skill should be used when the user asks to "create a slash command", "add a command", "write a custom command", "define command arguments", "use command frontmatter", "organize commands", "create command with file references", "interactive command", "use AskUserQuestion in command", or needs guidance on slash command structure, YAML frontmatter fields, dynamic arguments, bash execution in commands, user interaction patterns, or command development best practices for Claude Code.

57

Quality

66%

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/plugin-dev/skills/command-development/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

56%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 thorough, largely executable reference with strong examples and useful validation/troubleshooting patterns, but it overruns the token budget with padded intros and long sections that belong in its own reference files, and its progressive disclosure is inconsistent — a broken examples/ pointer and five unreferenced bundle files leave most of the bundle undiscoverable. Tightening the body to an overview plus quick-start patterns and linking every reference file would address the two weakest dimensions.

Suggestions

Move the Plugin-Specific Features, Integration with Plugin Components, and Validation Patterns sections into the existing reference files (or a new one) and keep one-paragraph summaries with pointers in SKILL.md, cutting the ~880-line body substantially.

Fix or remove the broken "see examples/ directory" pointer (the directory does not exist), and add clearly signaled links to the five unreferenced bundle files — especially interactive-commands.md, since the description triggers on "use AskUserQuestion in command" but the body never covers it.

Delete or compress generic padding Claude does not need: the "What is a Slash Command?" benefits list, the repeated statements of "commands are instructions for Claude", and the Overview key-concepts bullet list that restates the section headers.

DimensionReasoningScore

Conciseness

The body runs ~880 lines with several padded sections that assume too little of Claude: "What is a Slash Command?" with generic benefit bullets (Reusability, Consistency, Sharing, Efficiency), restatements of the same idea ("Commands are written for agent consumption, not human consumption" explained three ways), and long inline Plugin-Specific Features, Integration, and Validation sections that duplicate material the seven reference files exist to hold. This matches anchor 2 (noticeably verbose, several unnecessary explanations or padded sections) rather than anchor 3, because the padding is recurrent rather than incidental.

2 / 5

Actionability

Most guidance is copy-paste ready: complete command files with frontmatter (description, allowed-tools, argument-hint), usage invocations, and their expansions for $ARGUMENTS, positional args, @file references, !`bash`, and ${CLAUDE_PLUGIN_ROOT}. Minor gaps hold it at anchor 4 rather than 5: the core bash-execution syntax itself is deferred entirely to a reference, and the $IF($1,...) example appears with no syntax explanation.

4 / 5

Workflow Clarity

The reference-style progression is coherent (locations → file format → frontmatter → dynamic arguments → file references → bash execution → organization → patterns → troubleshooting), and the Validation Patterns section provides genuine feedback loops (test -f checks, "if build failed: analyze, suggest, provide steps"). It falls short of anchor 5 because there is no end-to-end sequence for actually creating and verifying a new command (write file → check /help → restart) — steps are scattered across Basics and Troubleshooting rather than sequenced.

4 / 5

Progressive Disclosure

References are one level deep and mostly well signaled (a dedicated pointer block at the end plus inline mentions of references/plugin-features-reference.md and references/frontmatter-reference.md, both real files), but the structure has real gaps: the final "see examples/ directory" points to a folder that does not exist, and 5 of the 7 bundle reference files (advanced-workflows, documentation-patterns, interactive-commands, marketplace-considerations, testing-strategies) are never referenced from the body — including interactive-commands.md, which covers the AskUserQuestion feature the description explicitly triggers on. Meanwhile ~300 lines of plugin/validation detail remain inline in SKILL.md. This lands on anchor 3: structure exists and references are signaled, but content that should be separate is inline and navigation to most of the bundle is missing.

3 / 5

Total

13

/

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 trigger-focused description with comprehensive natural trigger phrases and an unambiguous 'when', but it opens with a wall of quoted triggers and never explicitly states what the skill provides. Adding a brief leading capability sentence (e.g., "Guides creation and organization of Claude Code slash commands: frontmatter configuration, dynamic arguments, file references, and bash execution.") followed by the trigger list would lift specificity and completeness.

Suggestions

Lead with a third-person capability statement of what the skill provides (structure, frontmatter fields, dynamic arguments, file references, bash execution, organization) before the trigger list, so the 'what' is explicit rather than inferred from 'needs guidance on'.

Trim the quoted trigger list to the highest-probability natural phrases — the near-synonymous cluster ("create/add/write a custom command") adds length without adding trigger coverage, and verbosity is penalized.

Disambiguate from the skill-development skill by scoping the trigger terms, e.g. "for slash commands (.claude/commands/ markdown files)", to reduce conflict risk with the newer skills format.

DimensionReasoningScore

Specificity

The description names the domain (slash command development for Claude Code) and lists concrete trigger actions like "create a slash command", "define command arguments", and "use command frontmatter", but it never states what the skill itself provides or does — capabilities are only weakly implied via "needs guidance on slash command structure, YAML frontmatter fields, dynamic arguments, bash execution in commands". This sits between anchor 3 (domain and some concrete actions, not comprehensive as a capability statement) and anchor 4; the absent explicit 'what it does' keeps it at 3.

3 / 5

Completeness

The 'when' is exceptionally explicit ("This skill should be used when the user asks to..." plus a long trigger list), and a 'what' is present but only implicitly, as the topics the user "needs guidance on" (command structure, frontmatter fields, dynamic arguments, bash execution, interaction patterns, best practices) rather than a direct statement of what the skill does. It exceeds anchor 3 (which requires the 'when' to be missing) but falls short of anchor 5's concrete what+when pairing.

4 / 5

Trigger Term Quality

Trigger coverage is comprehensive and natural: "create a slash command", "add a command", "write a custom command", "define command arguments", "use command frontmatter", "organize commands", "interactive command", and "use AskUserQuestion in command", plus technical variants (YAML frontmatter, dynamic arguments, bash execution). This matches anchor 5's comprehensive natural-term coverage including synonyms; nothing common is missing.

5 / 5

Distinctiveness Conflict Risk

The niche is clear — Claude Code slash command development — with distinct triggers unlikely to fire for unrelated skills. Minor overlap remains with the sibling skill-development skill (the body itself notes the commands/ and skills/ formats "are loaded identically"), so a user asking to "create a skill" could plausibly hit either; that keeps it below anchor 5.

4 / 5

Total

16

/

20

Passed

Validation

87%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (885 lines); consider splitting into references/ and linking

Warning

frontmatter_unknown_keys

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

Warning

Total

14

/

16

Passed

Repository
anthropics/claude-plugins-official
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.