board
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill reads message board posts from various channels (e.g., dispatch, results) which serve as external, potentially untrusted sources of information. Malicious content within these posts could influence the agent's subsequent actions.
- Ingestion points: The
--read {channel}command inSKILL.mdtriggers the reading of markdown files from the file system into the agent's context. - Boundary markers: Absent. The instructions do not specify any delimiters or warnings to ignore instructions embedded within the board posts.
- Capability inventory: The skill possesses file read and write capabilities via
scripts/board_manager.py. - Sanitization: Absent. There is no evidence of content filtering or validation for the incoming message data.
- [COMMAND_EXECUTION]: The skill executes a local script
scripts/board_manager.pyusing the Python interpreter to perform its primary functions. While this is the intended design, it involves invoking shell commands with parameters (channel, author, message) that could be manipulated if not properly handled by the underlying script.
Audit Metadata