The Agent Skills Directory

[PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill is designed to ingest and process untrusted external data, including contracts, policies, and web-retrieved content.
Ingestion points: Processes data from the local file system using Read, Glob, and Grep tools, and external content via WebFetch and WebSearch tools.
Boundary markers: The instructions lack explicit boundary markers or directives to ignore instructions embedded within the processed legal documents or web data.
Capability inventory: The agent possesses powerful capabilities including Read, Write, Edit, WebFetch, and WebSearch, which could be exploited if an ingested document contains malicious instructions.
Sanitization: No sanitization or validation logic is defined for the external data before it is analyzed by the agent.

legal-advisor