The Agent Skills Directory

[SAFE]: The skill follows a structured workflow for task decomposition and project management, utilizing standard directory structures like .scratch for local output.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection by design, as it must ingest and process external content from URLs or files to generate tickets.
Ingestion points: External issue URLs, spec files, and conversation history (SKILL.md, Step 1).
Boundary markers: The instructions do not define specific delimiters for separating external data from agent instructions.
Capability inventory: The skill can write files to the local file system and interact with remote issue trackers (GitHub, Linear).
Sanitization: No specific content sanitization is described; however, the skill requires a mandatory user approval phase (Step 4) which serves as a primary security control.
[CREDENTIALS_UNSAFE]: The skill avoids hardcoded secrets by delegating configuration and credential management to a separate setup utility (/setup-matt-pocock-skills), which is a recommended security practice.

to-tickets