The Agent Skills Directory

[SAFE]: The skill functions as a read-only analyzer for project tracking. It does not attempt to exfiltrate data, execute remote code, or modify the system state in a malicious way. Use of local file system tools and git is consistent with its stated purpose.- [PROMPT_INJECTION]: Assessment of indirect prompt injection surface:
Ingestion points: Local design documents (e.g., design/gdd/*.md), sprint plans (production/sprints/), and git log output.
Boundary markers: Not present; the agent is instructed to read and interpret the contents directly.
Capability inventory: Includes Read, Glob, Grep, and Bash (used for git log operations).
Sanitization: No explicit sanitization or filtering of processed file content is defined.

scope-check