conductor-new-track
Pass
Audited by Gen Agent Trust Hub on Jul 7, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it interpolates unsanitized user input into project documentation and metadata files.
- Ingestion points: User-provided text for feature summaries, user stories, acceptance criteria, and technical notes within
SKILL.md. - Boundary markers: None identified; user input is directly inserted into Markdown and JSON templates without delimiters.
- Capability inventory: The skill performs multiple file-write operations to create specifications (
spec.md), implementation plans (plan.md), and metadata (metadata.json) within the project's directory structure. - Sanitization: No input validation or escaping mechanisms are present to prevent malicious instructions from being embedded in the generated documents, which could potentially influence the behavior of the agent when it later reads these files.
Audit Metadata