The Agent Skills Directory

[PROMPT_INJECTION]: The skill ingests untrusted data from the repository's code hunks via Git diffs. This creates a surface for indirect prompt injection, where malicious instructions embedded in the code could attempt to influence the agent's checklist generation.
Ingestion points: Git diff hunks retrieved in SKILL.md.
Boundary markers: Absent.
Capability inventory: Read-only access to the local Git repository.
Sanitization: No explicit filtering or sanitization of the diff content is specified.
[COMMAND_EXECUTION]: The skill executes git diff main...HEAD to retrieve code changes. This is a standard read-only operation intended for its primary purpose and uses a static command string.

preflight