bd-modify-product-description-and-tasks
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (HIGH): The skill's core function is to read and process external content from the 'bd' issue tracker (stored in
.beads/issues.jsonl). - Ingestion points:
bd show <issue-id>,bd create --file <path>, and the.beads/issues.jsonlfile itself. - Boundary markers: None are specified in the instructions to help the agent distinguish between issue data and instructions.
- Capability inventory: The skill allows for local command execution via the
bdCLI tool and modification of project files. - Sanitization: No sanitization or validation logic is present to filter malicious prompts embedded in issue fields.
- [Command Execution] (MEDIUM): The skill relies on the execution of the
bdCLI tool. While the commands used are specific (e.g.,bd create), the tool itself is an unverified dependency that must be pre-installed on the host system, and the agent's interaction with it provides a surface for potential command injection if parameters are not correctly escaped by the underlying execution environment.
Recommendations
- AI detected serious security threats
Audit Metadata