expansion-update
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill captures project status updates provided by the user and saves them into local YAML tracking files, which are retrieved and displayed in future interactions.
- Ingestion points: User-supplied progress descriptions provided during the project update step.
- Boundary markers: The skill does not employ specific delimiters when writing or reading back the stored status descriptions.
- Capability inventory: The skill's operations are limited to reading and writing project files within the specific path
~/.claude/plugins/config/claude-for-legal-zh/employment-legal/. - Sanitization: The skill stores the provided text directly into the YAML data structure without additional filtering.
- [DATA_EXPOSURE]: The skill interacts with files located in the
~/.claude/plugins/config/directory. These files contain operational project data such as task names, responsible parties, and deadlines specific to the user's legal expansion tracking, rather than sensitive system secrets or personal credentials.
Audit Metadata