The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external repositories, which creates a potential surface for indirect prompt injection if malicious instructions are embedded within the code or comments of the project being cleaned. 1. Ingestion points: Reads and evaluates files, project structure, and documentation in the target repository (SKILL.md). 2. Boundary markers: Absent. There are no explicit delimiters to separate instructions from the data being processed, although the skill uses guardrails to protect 'uncertain work'. 3. Capability inventory: The skill is authorized to perform file modifications and execute project-specific verification commands such as builds, tests, and linting (SKILL.md). 4. Sanitization: No specific sanitization or filtering of external file content is described before the content enters the agent's context.

housekeeper