agent-team-development
Pass
Audited by Gen Agent Trust Hub on Mar 11, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill facilitates an environment vulnerable to indirect prompt injection by design.
- Ingestion points: Untrusted data enters the agent context through implementation plan tasks passed to 'Implementer' subagents and through
git diffoutputs processed by the 'Reviewer' subagent. - Boundary markers: The skill lacks explicit delimiters (e.g., XML tags or unique markers) around the task text provided to subagents to distinguish instructions from potentially malicious data within the plan.
- Capability inventory: The spawned subagents have significant capabilities, including the ability to modify files, commit code, and execute shell commands for testing purposes.
- Sanitization: The skill instructions do not specify any validation, escaping, or filtering of the external plan content or code changes before they are interpolated into the subagent prompts.
Audit Metadata