The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill analyzes local repository content, including code and configuration files, which are external data sources that could potentially contain malicious instructions intended to mislead the agent. \n
Ingestion points: Read, Glob, and Grep tools are used in SKILL.md to process project files during the production review. \n
Capability inventory: The agent is granted Write, Edit, and WebFetch capabilities, which could be targets for manipulation via injected instructions. \n
Boundary markers: The skill does not explicitly define delimiters or instructions to ignore embedded prompts in the files it audits. \n
Sanitization: No content filtering or sanitization is specified for the data ingested from the repository.

attio-prod-checklist