sumsub-create-cross-check-preset
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it processes user-supplied configuration data and later summarizes the API's response, potentially re-injecting data into the agent's context.
- Ingestion points: User-defined specifications in JSON or YAML format are read via stdin and processed by the
scripts/build_cross_check_preset.pyscript. - Boundary markers: The instructions do not define specific markers or warning tags when the agent presents API-returned fields like 'title' or 'id' to the user.
- Capability inventory: The skill utilizes network access via
curlto reach Sumsub's API and executes local Python and Bash scripts for data processing and signing. - Sanitization: The
scripts/build_cross_check_preset.pyscript implements validation logic that checks for allowed enum values, data types, and length limits (e.g., 256 chars for title, 512 for description).
Audit Metadata