The Agent Skills Directory

[SAFE]: The skill instructions describe a legitimate developer workflow for managing code changes between local environments and source repositories. No malicious patterns or suspicious behaviors were detected in the command usage or process flow.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface by reading local file contents and git logs to identify changes. This is the primary intended function of the tool and is considered safe in this context. 1. Ingestion points: Reads contents from .agents/skills/ and skills-lock.json (SKILL.md). 2. Boundary markers: None identified. 3. Capability inventory: Uses git log, git diff, diff, and git commit (SKILL.md). 4. Sanitization: No specific sanitization of file content is requested before analysis.

backport-skill