The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to instructions embedded within the repository files it is designed to analyze.
Ingestion points: The skill uses the Read and Grep tools to ingest content from arbitrary files within the repository environment as defined in SKILL.md.
Boundary markers: The instructions lack explicit delimiters or warnings to the agent to treat content read from files as data rather than instructions, which could lead the agent to follow directions found inside the code or documentation.
Capability inventory: The skill's capabilities are limited by the allowed-tools frontmatter to Read, Grep, and Glob. It does not have permissions for network access, file modification, or arbitrary command execution.
Sanitization: No sanitization, filtering, or validation is performed on the data retrieved from the files before it is processed by the agent.

deep-research