{
  "markdown": "# enhance\n\nMaster enhancement orchestrator for plugins, agents, prompts, docs, hooks, and skills.\n\n## Overview\n\nThe enhance plugin provides specialized analyzers for different content types, identifying issues and suggesting improvements based on prompt engineering best practices.\n\n## Architecture\n\n```\n/enhance\n  -> /enhance:agent       - Agent-specific analysis (frontmatter, tool restrictions)\n  -> /enhance:prompt      - General prompt patterns (clarity, structure, examples)\n  -> /enhance:docs        - Documentation analysis (RAG optimization, readability)\n  -> /enhance:plugin      - Plugin structure (MCP tools, security patterns)\n  -> /enhance:claudemd    - Project memory optimization (CLAUDE.md/AGENTS.md)\n  -> /enhance:hooks       - Hook definitions (frontmatter, safety)\n  -> /enhance:skills      - SKILL.md structure and triggers\n  -> /enhance:cross-file  - Cross-file consistency (tools, agents, rules)\n```\n\n**Analysis depth**: Certainty-based findings (HIGH, MEDIUM, LOW)\n**Auto-fix**: Available for HIGH certainty issues with `--fix` flag\n**Model selection**: Opus for quality-critical analyzers, Sonnet for pattern-based checks\n\n## Commands\n\n### `/enhance`\n\nRun all applicable enhancers on current directory.\n\n```\n/enhance                    # Auto-detect and run all relevant analyzers\n/enhance --fix              # Apply HIGH certainty auto-fixes\n/enhance --verbose          # Include LOW certainty issues\n```\n\n### `/enhance:agent [target]`\n\nAnalyze agent prompt files for configuration and structure issues.\n\n```\n/enhance:agent                     # All agents in directory\n/enhance:agent my-agent.md         # Specific agent\n/enhance:agent --fix               # Apply auto-fixes\n```\n\n**Detects**: Missing frontmatter, unrestricted Bash, missing role section, tool configuration issues\n\n### `/enhance:prompt [target]`\n\nAnalyze prompts for prompt engineering best practices.\n\n```\n/enhance:prompt                    # All prompts in directory\n/enhance:prompt system-prompt.md   # Specific prompt\n/enhance:prompt --fix              # Apply auto-fixes\n```\n\n**Detects**: Vague instructions, missing examples, aggressive emphasis, structural issues, invalid code blocks (JSON/JS syntax, language mismatches, heading hierarchy)\n\n### `/enhance:docs [target]`\n\nAnalyze documentation for readability and RAG optimization.\n\n```\n/enhance:docs                      # All docs in directory\n/enhance:docs --ai                 # AI-only mode (aggressive optimization)\n/enhance:docs agent-docs/ --ai     # Specific directory\n```\n\n**Detects**: Verbose phrases, poor chunking, broken links, token inefficiency\n\n### `/enhance:plugin [target]`\n\nAnalyze plugin structure and MCP tool definitions.\n\n```\n/enhance:plugin                    # All plugins\n/enhance:plugin my-plugin          # Specific plugin\n/enhance:plugin --fix              # Apply auto-fixes\n```\n\n**Detects**: Missing schema fields, security patterns, version mismatches\n\n### `/enhance:claudemd`\n\nAnalyze project memory files (CLAUDE.md, AGENTS.md).\n\n```\n/enhance:claudemd                  # Find and analyze project memory\n/enhance:claudemd --fix            # Apply auto-fixes\n```\n\n**Detects**: Missing sections, broken references, README duplication, cross-platform issues\n\n### `/enhance:hooks`\n\nAnalyze hook definitions for frontmatter quality.\n\n```\n/enhance:hooks                     # All hook definitions\n/enhance:hooks pre-commit.md        # Specific hook\n```\n\n**Detects**: Missing frontmatter, missing name/description\n\n### `/enhance:skills`\n\nAnalyze SKILL.md files for required metadata and trigger clarity.\n\n```\n/enhance:skills                     # All SKILL.md files\n/enhance:skills enhance-docs         # Specific skill\n```\n\n**Detects**: Missing frontmatter, missing name/description, missing trigger phrase\n\n### `/enhance:cross-file`\n\nAnalyze cross-file semantic consistency across agents, skills, and commands.\n\n```\n/enhance:cross-file                 # Check cross-file consistency\n```\n\n**Detects**: Tools vs frontmatter mismatches, broken agent references, orphaned agents, skill tool mismatches, duplicate rules, contradictions\n\n## Agents\n\n| Agent | Purpose | Model |\n|-------|---------|-------|\n| `agent-enhancer` | Frontmatter, tool restrictions, agent structure | opus |\n| `prompt-enhancer` | Clarity, examples, structure, anti-patterns | opus |\n| `docs-enhancer` | RAG optimization, readability, token efficiency | opus |\n| `plugin-enhancer` | MCP schemas, security patterns, structure | sonnet |\n| `claudemd-enhancer` | Project memory validation, cross-platform | opus |\n| `hooks-enhancer` | Hook frontmatter, structure, safety | opus |\n| `skills-enhancer` | SKILL.md structure, trigger phrases | opus |\n| `cross-file-enhancer` | Cross-file semantic consistency (tools, agents, rules) | sonnet |\n\n## Certainty Levels\n\n| Level | Meaning | Auto-Fixable |\n|-------|---------|--------------|\n| HIGH | Definite issues | Some |\n| MEDIUM | Likely improvements | No |\n| LOW | Advisory suggestions | No |\n\nLOW certainty issues only shown with `--verbose` flag.\n\n## Common Flags\n\n| Flag | Description |\n|------|-------------|\n| `--fix` | Apply HIGH certainty auto-fixes |\n| `--verbose` | Include LOW certainty issues |\n| `--dry-run` | Show what would be fixed without applying |\n| `--ai` | AI-only mode (docs analyzer) |\n| `--both` | Both audiences mode (docs analyzer, default) |\n\n## Output Format\n\nEach analyzer generates a markdown report:\n\n```markdown\n## Analysis: {name}\n\n**File**: {path}\n**Analyzed**: {timestamp}\n\n### Summary\n- HIGH: {count} issues\n- MEDIUM: {count} issues\n- LOW: {count} issues (verbose only)\n\n### {Category} Issues ({n})\n\n| Issue | Fix | Certainty |\n|-------|-----|-----------|\n| Description | Suggested fix | HIGH |\n```\n\n## Integration\n\nCan be invoked by:\n- Direct command: `/enhance:*`\n- Phase 9 review loop during workflow\n- `delivery-validator` before shipping\n- Individual analysis workflows\n\n## Requirements\n\n- Claude Code\n- Node.js (for lib functions)\n\n## License\n\nMIT\n",
  "bytes": 5981,
  "sha": "a244bc111a13cd250979555edb9fd32165be3d7ef071ba01fa01e5b65b21ae9b",
  "repo_slug": "agent-sh/enhance",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_agent_sh_enhance_enhance_ead18f9b/readme"
}