bmad-idea
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates workflows like brainstorming, design thinking, and storytelling which involve processing user-provided text and potentially reading existing project files. This creates a surface for indirect prompt injection if malicious instructions are embedded in the data being processed.
- Ingestion points: Processes user-provided creative prompts and reads project files via
Read,Grep, andGlobtools during ideation and analysis workflows (e.g., inbmad-cis-brainstormingorbmad-cis-problem-solving). - Boundary markers: There are no specified delimiters or "ignore embedded instructions" warnings in the agent's prompt definitions to distinguish between data and commands.
- Capability inventory: The skill is configured with access to
Read,Write,Bash,Grep, andGlobtools, enabling file operations and shell command execution. - Sanitization: No sanitization or validation mechanisms for external content are documented in the skill's instructions.
- [EXTERNAL_DOWNLOADS]: The documentation references external sources for installation and technical reference.
- Evidence: Provides an installation command using
https://github.com/supercent-io/skills-templateand references upstream documentation athttps://github.com/bmad-code-org/bmad-module-creative-intelligence-suite.
Audit Metadata