The Agent Skills Directory

[PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill is designed to ingest and process content from a project's codebase, including instruction files such as AGENTS.md, CLAUDE.md, and GEMINI.md, to generate documentation. This creates a surface where malicious instructions embedded in the project files could influence the agent's behavior during the README generation process.
Ingestion points: The skill reads various files within the repository during the project discovery phase, as specified in rules/project-discovery.md and rules/project-discovery.ko.md.
Boundary markers: The instructions do not define explicit boundary markers or directives to ignore instructions contained within the ingested project files, although they emphasize grounding output in repository evidence.
Capability inventory: The skill utilizes read, edit, write, find, grep, and ls tools to interact with the file system.
Sanitization: No specific sanitization or filtering of the ingested file content is mentioned before it is processed by the model.

readme-maker