From skill-maker
Performs a full quality review of a Claude Code skill, then automatically fixes detectable issues in frontmatter, body, and structure. Reports changes and re-validates.
How this command is triggered — by the user, by Claude, or both
Slash command
/skill-maker:fix path to skill folder or SKILL.mdThe summary Claude sees in its command listing — used to decide when to auto-load this command
# Auto-Fix Skill Issues Automatically review and fix quality issues in an existing skill. Runs a full review first, then applies fixes. ## Input The user provides: `$ARGUMENTS` If `$ARGUMENTS` is a path, use it directly. If `$ARGUMENTS` is empty, ask: "Which skill should I fix? Provide the path to the skill folder or SKILL.md file." ## Process ### 1. Run Review Perform the same comprehensive review as `/skill-maker:review`: - Read SKILL.md and all supporting files - Assess structure, frontmatter, description, body, triggering ### 2. Identify Fixable Issues Categorize findings: **A...
Automatically review and fix quality issues in an existing skill. Runs a full review first, then applies fixes.
The user provides: $ARGUMENTS
If $ARGUMENTS is a path, use it directly.
If $ARGUMENTS is empty, ask: "Which skill should I fix? Provide the path to the skill folder or SKILL.md file."
Perform the same comprehensive review as /skill-maker:review:
Categorize findings:
Auto-fixable:
--- delimiters in frontmatterRequires user input:
For each auto-fixable issue:
For issues requiring user input:
After all fixes applied, show:
FIXES APPLIED: [skill-name]
═══════════════════════════
Fixed:
- [Fix 1]: [what was changed]
- [Fix 2]: [what was changed]
- [Fix 3]: [what was changed]
Skipped (needs user input):
- [Issue 1]: [why it needs input]
Before: Grade [X] → After: Grade [Y]
Run the full validation again to confirm all fixes worked. Present the updated scores.
--- delimitersnpx claudepluginhub vcode-sh/vibe-tools --plugin skill-maker/heal-skillDetects and fixes incorrect SKILL.md files by analyzing conversation context, proposing corrections, and applying changes with user approval. Supports optional commit.
/heal-skillAnalyzes conversation context to detect which skill is running, identifies incorrect instructions or outdated API references in SKILL.md, and applies approved corrections across related files.
/skill-improverIteratively reviews and fixes a Claude Code skill until it meets quality standards. Accepts a skill name or path and optional --max-iterations flag.
/audit-skillAudits Claude Code SKILL.md files for structure, description quality, content effectiveness, and progressive disclosure. Supports --full and --mini depth flags for thorough or quick reviews.
/corgi-improve-skillRefines or creates an agent skill from conversation evidence. Accepts a skill path and gap description; edits existing skills, scaffolds new ones, or summarizes gaps without changes.
/heal-skillUpdates a skill's SKILL.md and related files based on corrections discovered during execution, with user approval and optional git commit.