roadmap-task-flow
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to capture and enrich tasks from free-form user input, which creates a potential surface for indirect prompt injection if the ingested data contains malicious instructions meant to influence the agent's behavior.
- Ingestion points:
SKILL.mdandreferences/workflow-boundaries.mdspecify capturing tasks from "free-form user input". - Boundary markers: No explicit delimiters or instructions are provided to the agent to disregard embedded instructions within task descriptions.
- Capability inventory: The skill is limited to project management tools (
create_task,update_task,list_tasks). It lacks high-risk capabilities such as arbitrary shell execution, external network requests, or sensitive file system access. - Sanitization: No explicit sanitization or validation steps are described for handling external user input.
Audit Metadata