gds-quick-dev

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local shell commands to facilitate development tasks. This includes running a customization resolver script (resolve_customization.py) from the project's internal _bmad directory, using git for version control, and launching the code editor. These actions are consistent with the skill's intended purpose.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided intent and external planning artifacts to generate code and specifications, creating a surface for injection.\n
  • Ingestion points: User intent and planning documents (PRD, architecture docs) in the {planning_artifacts} directory.\n
  • Boundary markers: The skill uses a structured markdown template and a mandatory human approval checkpoint before implementation begins.\n
  • Capability inventory: The agent can execute shell commands, write files to the project, and spawn sub-agents for specialized tasks.\n
  • Sanitization: The instructions direct the agent to ignore directives within the intent that attempt to bypass the defined workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 08:01 PM
Security Audit — agent-trust-hub — gds-quick-dev