policy-redraft
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill reads from a local configuration file located at
~/.claude/plugins/config/claude-for-legal-zh/regulatory-legal/CLAUDE.md. Accessing hidden configuration directories within the user's home folder is a sensitive operation. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external data that could contain malicious instructions designed to influence the agent's behavior.
- Ingestion points: The skill explicitly requests "Current policy text" (1b) and "Regulation text" (1c) from user messages or file paths.
- Boundary markers: The instructions do not define any delimiters or explicit commands to the agent to ignore instructions embedded within the provided legal or policy texts.
- Capability inventory: The skill has the capability to read from the local filesystem and write new markdown files containing policy redrafts.
- Sanitization: There is no evidence of sanitization, filtering, or validation performed on the ingested text before it is processed by the model.
Audit Metadata