The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes project source code and markdown files to generate and validate documentation. This creates an attack surface where untrusted data from these files could potentially contain malicious instructions targeting the AI agent.
Ingestion points: Processes various project files including **/*.md, **/*.py, **/*.ts, package.json, and setup.py as defined in references/validations.md.
Boundary markers: No explicit delimiters are specified to separate untrusted file content from the agent's system instructions.
Capability inventory: The skill is designed for technical writing and documentation validation, utilizing the agent's underlying capabilities to read files and generate text.
Sanitization: The skill does not implement specific sanitization logic for the content it reads, although it does include patterns to detect and flag potentially unsafe content like hardcoded secrets.

docs-engineer