proposition-checking
Pass
Audited by Gen Agent Trust Hub on Jul 8, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXPOSURE]: The skill operates in a 'Live source mode' that utilizes web search or browser tools to retrieve legal authorities. Example configurations point to well-known, official government repositories such as the UK National Archives (caselaw.nationalarchives.gov.uk).
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external data, including user-uploaded legal drafts and third-party web content. It mitigates the risk of indirect prompt injection by enforcing a structured reporting format (proposition vs. source excerpt) and including explicit instructions to separate the assertion from the supporting text.
- [COMMAND_EXECUTION]: The skill instructions emphasize the use of configured environment tools (like MCP or APIs) for retrieval but do not include any arbitrary shell command execution, persistence mechanisms, or privilege escalation attempts.
- [DATA_EXPOSURE]: The skill acknowledges the handling of sensitive legal materials (privileged/confidential drafts) and includes specific warnings for the user to utilize approved, secure processing environments.
Audit Metadata