The Agent Skills Directory

[SAFE]: The skill focuses on organizing and querying documentation within the docs/features/ directory. It relies on standard tools like Read, Grep, and Glob to interact with files.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface by reading markdown files from the project directory. However, because the skill has restricted tool access and no network capabilities, this surface does not present a significant security risk.
Ingestion points: Reads request documents and templates from the docs/features/ directory (SKILL.md, operations.md).
Boundary markers: None explicitly defined in instructions, but content is structured markdown.
Capability inventory: Limited to file reading and searching; no network access, arbitrary command execution, or sensitive file access.
Sanitization: Not applicable as the skill primarily structures internal documentation.

request-tracking