BookForge Skills is a library of agent skills distilled from non-fiction books. Skills are static Markdown files that instruct AI agents — they do not execute code in the skill library itself, and the library has no runtime components.

If you discover a concern related to the content of a skill — for example, a skill that instructs an agent to produce harmful output, leak sensitive information, or violate common safety norms — please report it by opening a GitHub issue or emailing the address in TAKEDOWN.md.

For concerns about the BookForge pipeline itself (the code that generates skills), please report them to the bookforge repository instead.

• A link to the specific skill file or directory
• A clear description of the concern
• An example of the problematic output, if applicable
• Any suggested remediation

• We acknowledge reports within 48 hours of receipt.
• We investigate and respond with an action plan within 7 days.
• Critical concerns are addressed as quickly as possible, often within 48 hours of verification.
• We maintain a private compliance log of every report and its resolution.

This policy applies to the content and structure of skills in this repository. It does not apply to:

• The underlying AI model you use with the skill (report those to the model provider)
• The agent platform you install the skill into (report those to the platform)
• Third-party tools that BookForge skills reference

• TAKEDOWN.md — DMCA and rights-holder takedown procedure
• COPYRIGHT.md — legal framework and attribution practice
• CODE_OF_CONDUCT.md — contributor conduct expectations