Automatically fixes mechanical issues found during plugin validation, including JSON formatting, missing shebangs, file permissions, hook types, frontmatter gaps, and settings.json defaults. ALWAYS asks for confirmation before modifying any files. [EXPLICIT] Trigger: fix issues, auto-fix, fix common problems, repair plugin. [EXPLICIT]
From jm-adknpx claudepluginhub javimontano/jm-adk-alfaThis skill is limited to using the following tools:
agents/guardian.mdagents/lead.mdagents/specialist.mdagents/support.mdevals/evals.jsonknowledge/body-of-knowledge.mdknowledge/knowledge-graph.mdprompts/meta.mdprompts/primary.mdprompts/variations/deep.mdprompts/variations/quick.mdreferences/auto-fix-catalog.mdtemplates/output.docx.mdtemplates/output.html"Automate the mechanical so humans can focus on the meaningful."
Auto-remediates mechanical issues found during plugin validation: JSON syntax, missing shebangs, script permissions, incompatible hook types, missing frontmatter fields, and settings.json defaults. Presents all proposed changes for user approval before modifying any file. [EXPLICIT]
Inventory fixable issues -- review findings from previous validation results. If no validation has been run yet, inform the user and suggest running validations first. Categorize issues into fixable (mechanical) vs. non-fixable (requires human judgment). List only mechanical issues for auto-fix. [EXPLICIT]
Detect and plan JSON trailing comma fixes -- scan all .json files for trailing commas (comma before ] or }). Plan: parse the JSON, reformat with standard 2-space indentation, write clean JSON back. [EXPLICIT]
Detect and plan missing shebang fixes -- scan all .sh files for missing #!/bin/bash (or #!/usr/bin/env bash) on line 1. Plan: prepend #!/bin/bash followed by a newline. [EXPLICIT]
Detect and plan non-executable script fixes -- check all .sh files in scripts/ and hooks/ for missing execute permission. Plan: mark for chmod +x. [EXPLICIT]
Detect and plan prompt-to-command hook conversion -- scan hooks/hooks.json for entries using type: "prompt" or type: "agent" on non-ToolUseContext events (any event other than PreToolUse, PostToolUse, PermissionRequest). Plan: convert to type: "command" using echo as the command, preserve the original prompt text as a comment in the command string. [EXPLICIT]
Detect and plan missing frontmatter field fixes -- scan all SKILL.md files for missing required frontmatter fields (name, description, allowed-tools). Plan:
name: derive from parent directory name in kebab-casedescription: add placeholder "TODO: Add description with trigger phrases"allowed-tools: add default [Read]Detect and plan settings.json agent key fix -- check if settings.json exists but lacks an "agent" key. Plan: find the first agent directory under agents/, set "agent" to that path. [EXPLICIT]
Present all proposed changes for confirmation -- display a numbered list of every planned modification:
STOP HERE AND WAIT FOR USER APPROVAL. Do not modify any files until the user explicitly confirms. The user may approve all, approve selectively by number, or reject. [EXPLICIT]
Apply approved fixes -- execute only the changes the user approved. For each fix:
Re-validate after fixes -- run the relevant validation checks on modified files to confirm the issues are resolved. Report any fixes that did not resolve their target issue. [EXPLICIT]
Generate fix report -- produce a summary listing:
Bad fix proposal:
Fix 1: Fix hooks.json
Missing: no file path, no before/after, no change type. [EXPLICIT]
Good fix proposal:
Fix 1 [Hook Type] hooks/hooks.json
Before: { "type": "prompt", "prompt": "Review safety" } on SessionStart
After: { "type": "command", "command": "echo 'Review safety'" } on SessionStart
Reason: type:prompt requires ToolUseContext; SessionStart does not provide it. [EXPLICIT]
Includes: fix number, category, file path, before/after with context, reason for the change. [EXPLICIT]
.sh files in scripts/ and hooks/ should be modified. [EXPLICIT]Example invocations:
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.