skills/agentskillexchange/skills/Coordinate software delivery roles with MetaGPT/Gen Agent Trust Hub
Coordinate software delivery roles with MetaGPT
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted product requirements and optional repository context to generate software development artifacts, creating an attack surface for indirect prompt injection.\n
- Ingestion points: User-provided product requirements and repository context mentioned in SKILL.md.\n
- Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore potentially malicious embedded instructions in the ingested text.\n
- Capability inventory: The skill leverages MetaGPT to produce coordinated product, architecture, and engineering outputs, including code generation.\n
- Sanitization: No evidence of validation or sanitization of input requirements is provided in the skill instructions.\n- [EXTERNAL_DOWNLOADS]: The skill recommends installing the 'metagpt' package via pip and requires 'node' and 'pnpm' as prerequisites. These are standard dependencies for the framework obtained from public repositories.
Audit Metadata