igrantio-credential-schema-sca-payment-account
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill directs the agent to download JSON schema definitions and metadata from an external GitHub repository not associated with the primary vendor.
- Evidence: URLs in the 'Source is the registry' section point to
raw.githubusercontent.com/decentralised-dataexchange/verifiable-data-registry/to fetch the authoritative schema files. - [INDIRECT_PROMPT_INJECTION]: The skill establishes an interaction pattern that gathers untrusted user input to be used in sensitive API configurations.
- Ingestion points: User/integrator responses to the eight configuration questions (Environment, API key, Format, Claims, Issuance mode, Revocation, Signing, and Trust list) in the 'Ask the integrator first' section.
- Boundary markers: None. The skill does not suggest the use of delimiters or provide instructions to the agent to treat these inputs as literal text or data only.
- Capability inventory: The skill uses the gathered data to build a credential definition and perform a
POSTrequest to the iGrant.io API via theigrantio-api-issuertool. - Sanitization: None. The instructions lack logic for validating, escaping, or sanitizing the input provided by the integrator before it is used to generate the API request payload.
Audit Metadata