Create exceptional, deterministic Claude skills that guide AI behavior with precision. Use this skill when the user asks to create new skills, document capabilities, or codify expertise into reusable instructions. Generates structured skill files that produce consistent, high-quality outputs.
/plugin marketplace add mahidalhan/skilled-spec/plugin install skill-tools@skilled-intelligenceThis skill inherits all available tools. When active, it can use any tool Claude has access to.
LICENSE.txtscripts/init_skill.pyscripts/package_skill.pyscripts/quick_validate.pyThis skill guides creation of exceptional Claude skills that avoid vague "be helpful" platitudes. Write deterministic instructions that produce consistent, high-quality outputs regardless of how many times they're invoked.
The user provides skill requirements: a domain to codify, an expertise to capture, a workflow to standardize, or a quality bar to enforce. They may include context about the target audience, existing patterns, or specific anti-patterns to avoid.
Before writing any instructions, understand the expertise and commit to a DETERMINISTIC structure:
CRITICAL: Skills must be deterministic, not aspirational. Every instruction should constrain output. If two different Claude instances reading your skill could produce wildly different results, the skill has failed. Specificity over generality. Always.
Then write a skill file that is:
Focus on:
name (kebab-case), description (when to use, what it produces), license. Description should contain trigger phrases ("when the user asks to...").**Concept**: Question? Elaboration. This forces planning before execution.references/ folder with concrete examples. Keep SKILL.md lean—reference files hold the detail..claude/commands/{skill-name}.md) with 2-3 lines: title, one-line description, and Uses @{skill-name} skill.NEVER write vague instructions ("be thorough", "consider carefully"), omit the CRITICAL callout, exceed 45 lines, use more than 5 Focus bullets, create skills without naming specific anti-patterns, write instructions that don't constrain output ("you may choose to..."), or ship a skill without its companion command file.
Adapt skill complexity to domain scope. Narrow skills (diagram types, test patterns) need concrete examples. Broad skills (design, architecture) need decision frameworks. Meta-skills (like this one) need structural templates.
IMPORTANT: Match skill tone to domain culture. Technical skills use precise language. Creative skills permit more latitude. Process skills emphasize sequence. The skill should feel native to its domain, not imposed from outside.
Remember: Claude is capable of extraordinary instruction-following. Don't write weak suggestions—write strong constraints that unlock consistent excellence across every invocation.
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.
This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent colors", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins.
This skill should be used when the user asks to "create a hook", "add a PreToolUse/PostToolUse/Stop hook", "validate tool use", "implement prompt-based hooks", "use ${CLAUDE_PLUGIN_ROOT}", "set up event-driven automation", "block dangerous commands", or mentions hook events (PreToolUse, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification). Provides comprehensive guidance for creating and implementing Claude Code plugin hooks with focus on advanced prompt-based hooks API.