building-workflows

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the creation of workflows that process external data such as events and person properties, which serves as an ingestion point for untrusted data.
  • Ingestion points: Data enters the context through workflows-test-run (globals) and workflows-get-invocation (triggering payloads) as described in SKILL.md and references/lifecycle-and-debugging.md.
  • Boundary markers: The skill does not define explicit delimiters for variable interpolation within the workflow instructions.
  • Capability inventory: Workflows have the capability to execute side effects via function, function_email, and function_push nodes, which include network and communication operations.
  • Sanitization: While the skill emphasizes structural validation of the JSON graph, it does not provide specific instructions for sanitizing the values processed within the workflow steps.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 05:15 PM
Security Audit — agent-trust-hub — building-workflows