The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the repository (such as README.md and existing instruction files) to create or update AGENTS.md and CLAUDE.md, which are then used to guide agent behavior in every subsequent session.
Ingestion points: Reads repository metadata, README.md contents, and existing instruction files (SKILL.md).
Boundary markers: No delimiters or protective instructions are used when interpolating external content into the generated instruction files.
Capability inventory: The skill has the ability to write files to the repository and execute shell commands (SKILL.md).
Sanitization: There is no evidence of sanitization or filtering of the ingested content before it is added to the agent's persistent instruction files.
[COMMAND_EXECUTION]: The skill directs the agent to identify and execute shell commands found within the repository's build files, scripts, and CI configurations to verify they work.
Evidence: Found in SKILL.md under the 'Create' section: 'fill Commands with the commands the build files, scripts and CI actually define... Run the cheap ones to confirm they work.'

agents-md