The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute various shell commands to interact with the local repository.
Evidence: Suggests using node scripts/ci/catalog.js, find, rg, and npm commands to discover components, search files, and manage installations within the project workspace.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from repository files which could theoretically contain instructions intended to influence the agent's behavior.
Ingestion points: Reads from SKILL.md, README.md, agent.yaml, and command outputs from rg search results.
Boundary markers: No explicit delimiters or instructions are provided to the agent to treat the file content strictly as data.
Capability inventory: The skill utilizes shell command execution (node, npm, find, rg) and file system access.
Sanitization: There is no mention of sanitizing or validating the content retrieved from repository files before it is processed by the agent.

ecc-guide