suprsend-workflow-schema
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill documents a workflow system designed to ingest and process untrusted data from external sources, creating a surface for indirect prompt injection.
- Ingestion points: Data enters the system via workflow trigger payloads (SKILL.md) and response bodies from external API calls (node-fetch.md, node-webhook.md).
- Boundary markers: The documentation does not specify the use of delimiters or instructions to ignore embedded commands within the processed data.
- Capability inventory: The system supports network operations (httpapi_fetch, httpapi_webhook), user profile modifications (userupdate), and management of user lists and object subscriptions.
- Sanitization: There is no explicit mention of data sanitization or validation logic for external content before it is used in subsequent workflow steps.
- [DYNAMIC_EXECUTION]: The workflow schema documentation describes the use of JSONNET and Handlebars for runtime evaluation of variables, URL construction, and request body generation across several node types, including 'transform', 'httpapi_fetch', 'httpapi_webhook', and 'userupdate'.
- [DATA_EXFILTRATION]: The skill defines nodes ('httpapi_fetch' and 'httpapi_webhook') that allow workflows to perform network requests to external URLs, supporting various HTTP methods (GET, POST, PUT, PATCH, DELETE) to send or retrieve data.
Audit Metadata