context-driven-development
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a set of guidelines and a structured directory layout for maintaining project context. It defines how product goals, technical stacks, and workflows should be documented to assist AI development.
- [INDIRECT_PROMPT_INJECTION]: The skill describes an attack surface by instructing the agent to ingest and follow instructions from various project files (e.g.,
product.md,workflow.md,tech-stack.md). - Ingestion points: Files located in the
conductor/directory, includingproduct.md,tech-stack.md,workflow.md,tracks.md, and language-specific style guides. - Boundary markers: None explicitly defined in the skill instructions; however, the agent is directed to "Verify context accuracy with stakeholders."
- Capability inventory: The skill mentions internal agent commands like
/conductor:setupfor artifact creation and analysis but does not define any shell execution or network tools within the SKILL.md itself. - Sanitization: No specific sanitization or filtering of the content within these context files is mentioned.
Audit Metadata