bmad-index-docs
Pass
Audited by Gen Agent Trust Hub on Mar 21, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes the content of arbitrary files within a user-provided directory.
- Ingestion points: Step 3 (Generate Descriptions) involves the agent reading the actual content of all files in the target folder.
- Boundary markers: No explicit delimiters or system instructions are provided to the agent to disregard potential commands or role-play instructions embedded within the documents being indexed.
- Capability inventory: The skill possesses the capability to scan directories, read file data, and write to the local filesystem (index.md).
- Sanitization: No evidence of sanitization or validation of the content read from files exists before the text is processed by the model for description generation.
Audit Metadata