The Agent Skills Directory

[SAFE]: The skill implements a structured workflow for legal documentation, including validation steps to ensure case identifiers exist before modification.
[DATA_EXPOSURE]: The skill reads from and writes to specific paths under ~/.claude/plugins/config/claude-for-legal-zh/. This is standard behavior for a plugin-based skill and does not involve accessing sensitive system credentials or exfiltrating data to external servers.
[INDIRECT_PROMPT_INJECTION]: While the skill appends user-provided text (summaries and event descriptions) to history files, it does not process this data in a way that triggers code execution or influences agent logic beyond formatting. The skill also includes specific procedural gatekeeping, such as requiring lawyer consultation for settlement acceptance, which functions as a domain-specific safety check.

matter-update