igrantio-consent-records
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill provides a standard Node.js/TypeScript implementation for consent management without any malicious intent or hidden code.
- [DATA_EXFILTRATION]: Network operations are directed to the vendor's official API endpoints (demo-api.igrant.io and staging-api.igrant.io). Secret handling instructions, such as storing the OWS_API_KEY in a .env file, align with standard secure development practices and prevent accidental exposure.
- [INDIRECT_PROMPT_INJECTION]: The skill handles user-controlled parameters like dataAgreementId and opt-in status that are transmitted to an external API. * Ingestion points: Data enters via URL parameters and JSON bodies in
references/src/consentRouter.ts. * Boundary markers: The implementation acts as a traditional backend service rather than a text-prompt generation skill, reducing the surface for prompt-based attacks. * Capability inventory: Capabilities are limited to network operations viafetchinreferences/src/consentClient.ts. * Sanitization: Inputs are correctly sanitized usingURLSearchParamsfor query strings and standardJSON.stringifyfor request bodies. - [COMMAND_EXECUTION]: Development and build scripts in
package.jsonutilizetsxandtypescript, which are standard industry tools for running and compiling TypeScript code. No arbitrary or dangerous shell command execution patterns were found.
Audit Metadata