The Agent Skills Directory

[PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill is designed to ingest and analyze untrusted contract text which could contain malicious instructions meant to manipulate the agent's behavior.\n- Ingestion points: User-provided contract text is ingested in the 'Contract intake' workflow step within SKILL.md.\n- Boundary markers: The skill does not use specific delimiters or instructions to treat the ingested contract text as data only, increasing the risk that the agent may follow instructions embedded within the document.\n- Capability inventory: Analysis of the skill reveals no capabilities for network operations, file system modifications, or arbitrary code execution.\n- Sanitization: The skill does not implement any validation, escaping, or filtering of the external contract content before processing.

contract-review