Validate skill definitions against agentskills spec

Validates SKILL.md files against the agentskills specification and Anthropic best practices in a single read-only pass, catching structural errors, semantic contradictions, naming anti-patterns, and quality gaps. It parses the target file's YAML frontmatter, then runs the Free-tier checks across five categories - Structure (name format, WHAT+WHEN description shape, allowed-tools, categories, XML injection), Body (line count, hardcoded paths, stale dates, empty sections, deprecated syntax, MCP tool qualification), Naming (vague terms, single-word names, gerund-form suggestions), Semantic (contradictions, ambiguous terms, missing output format, wisdom or platitude language, misplaced triggers), and Quality (examples, error handling, triggers, output format, prerequisites, negative triggers) - then calculates an overall 0-100 score with penalties of -20 for a critical, -5 for a warning, and -1 for a suggestion, with suggestions capped at -15 total. It returns a structured report: score, letter grade (Excellent, Good, Needs Work, or Poor), and an issue list with check IDs, line numbers, messages, and fix suggestions.

Use it when the user says "check skill," "skillcheck," or "validate SKILL.md," when reviewing a skill before publishing it to a marketplace, when debugging why a skill isn't triggering correctly, or when onboarding a team to skill-authoring standards. Do not use it for anti-slop detection, security scanning, or token analysis - those require the paid SkillCheck Pro product. It is read-only and never modifies files, validates only the SKILL.md content itself rather than referenced files or scripts, checks a single file at a time without cross-checking other skills in the same directory, and its semantic checks (contradiction detection, wisdom/platitude flags) are heuristic with roughly a 5% false-positive rate.

Input is a target SKILL.md file. Output is a structured score report, for example:

### weekly-report Check Results [FREE]

Score: 85/100 (Good)

### Warnings (2)
  - 1.2-desc-when (line 3): Description missing WHEN clause
  - 4.5-desc-no-triggers (line 3): Description lacks triggering conditions

### Suggestions (1)
  - 3.4-gerund-naming (line 2): Skill name could use gerund form

### Passed Checks: 28

A clean skill returns a perfect score with all checks passed and no issues listed. Each issue is tagged with a check ID (like 1.2-desc-when) so the exact rule can be looked up, and re-running after fixes confirms the score improved.

Reads only the local SKILL.md file and its YAML frontmatter - no external services or network calls in the Free tier. Skips code blocks and inline code when scanning for ambiguous-term false positives, so wrapping a legitimate term in backticks avoids a false flag; more advanced checks (anti-slop, security, WCAG, token, enterprise, workflow) are handled by the separate paid SkillCheck Pro product rather than this tool.

Skill authors and teams who want an automated, repeatable pre-publish check on a SKILL.md file's structure, naming, semantics, and quality before submitting it to a marketplace or onboarding others to author skills the same way.