igrantio-credential-schema-qesac
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to fetch and prioritize external data from a GitHub repository to ensure the registry's content takes precedence over the skill's instructions ("the registry wins"). This creates a surface where malicious content in the remote registry could influence the agent's behavior.
- Ingestion points: The agent is directed to fetch raw JSON schema and metadata documents from the
decentralised-dataexchange/verifiable-data-registryrepository. - Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are provided for the external content.
- Capability inventory: The skill provides templates for credential definitions and references the
igrantio-api-issuerskill for API interactions. - Sanitization: No sanitization or validation logic is specified for the fetched remote content.
- [EXTERNAL_DOWNLOADS]: The skill references external JSON schema files and documentation hosted on GitHub.
- Source: Resources are fetched from
raw.githubusercontent.com/decentralised-dataexchange/verifiable-data-registry/. - Context: These downloads are used to retrieve the latest source of truth for credential schema definitions.
Audit Metadata