frontmatter

Stamp and scan AI provenance metadata blocks in source files.

Updated Dec 19, 2025
One-click install
npx skills add https://github.com/a3lem/my-claude-plugins --skill frontmatter-a3lem
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: frontmatter
Source: https://github.com/a3lem/my-claude-plugins/tree/main/plugins/frontmatter/skills/frontmatter
Command: npx skills add https://github.com/a3lem/my-claude-plugins --skill frontmatter-a3lem

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill addresses the challenge of tracking AI-generated or AI-assisted code, ensuring clarity on human review status, access controls, and the specific rules or skills that governed AI modifications.

Core Features & Use Cases

  • AI Provenance Tracking: Stamp files with metadata indicating human review status, access permissions (read/write/hidden), and associated rules or skills.
  • Access Control: Enforce read-only or hidden access for AI agents on sensitive files.
  • Use Case: When an AI modifies a file, it automatically resets the human-reviewed flag to false, prompting a human review. It can also stamp new files with access = "read" to prevent accidental AI writes.

Quick Start

Stamp the file src/main.py with human-reviewed = false and access = "read".

Frequently Asked Questions about frontmatter

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I track AI provenance metadata in source files?

AI provenance metadata is tracked by stamping source files with a custom comment-based block that records human review status, access permissions, and rule references, ensuring clarity on how AI modified the code.

How do I mark AI-generated code as requiring human review?

To mark AI-generated code for human review, stamp the file with a metadata block setting human-reviewed to false. This flag automatically resets whenever an AI agent modifies the file.

Can I enforce read-only access control for AI agents on sensitive files?

Yes, you can enforce read-only access for AI agents by stamping files with the access = "read" metadata field, preventing accidental AI writes on sensitive code.

What is the best way to scan repository file coverage for AI metadata?

The best way to scan repository file coverage for AI metadata is using automated scanning to parse custom comment-based provenance blocks across the codebase, reporting which files contain review status and access control stamps.

Do I need specific dependencies to manage AI provenance metadata?

No specific dependencies are required to manage AI provenance metadata. The Skill uses standalone scripts to stamp and scan custom comment-based blocks directly within source files.