long-term-plan
Pass
Audited by Gen Agent Trust Hub on Aug 21, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill instructions define a workflow for ingesting untrusted user input and using it to generate project tasks, creating an indirect prompt injection surface.
- Ingestion points: Stage 1 prompts the agent to collect goals, timeframes, and constraints directly from user dialogue.
- Boundary markers: No delimiters are defined in SKILL.md to separate user-provided content from agent instructions when calling tools.
- Capability inventory: The skill utilizes tools like create_tasks and update_plan_document to generate tasks that the agent is instructed to execute automatically (auto tasks).
- Sanitization: There is no mention of validation, filtering, or escaping of user-provided strings prior to their use in tool execution.
Audit Metadata