agently-triggerflow-orchestration
Pass
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface in its orchestration patterns. Ingestion points: AI-generated outputs in examples/iterative_refinement_loop.py and examples/plan_execute_basic.py are fed back into agent inputs. Boundary markers: No delimiters are used to separate the instructions from the potentially untrusted AI-generated content. Capability inventory: The framework allows the agent to generate and process structured data in loops. Sanitization: There is no validation or escaping of the intermediate outputs before they are interpolated into new prompts.
Audit Metadata