specification-writing
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to execute or interact with a local script named
reviewer_route.pyas part of its reviewer selection process and makes use of theBashtool for operations. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from
$ARGUMENTSand external files likepatent/CLAIMS.mdandpatent/INVENTION_DISCLOSURE.mdwhich are then interpolated into prompts for the agent's workflow and secondary review sessions. 1. Ingestion points: Data is ingested from user arguments and files includingCLAIMS.mdandINVENTION_DISCLOSURE.md. 2. Boundary markers: No delimiters or specific instructions to ignore embedded commands are specified for the ingested content. 3. Capability inventory: The skill has access toBash,Write, andWebSearchtools. 4. Sanitization: No sanitization or validation of the input content is mentioned. - [DYNAMIC_EXECUTION]: The skill uses a dynamic mechanism to resolve reviewer configurations from
.aris/reviewer.jsonand utilizes thespawn_agentdirective to create new execution contexts with specific model parameters.
Audit Metadata