doc-req-autopilot
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill automates the document workflow by executing several local Python and shell scripts (e.g.,
validate_sys.py,validate_req_template.sh,validate_req_spec_readiness.py) to verify document structure and readiness scores within the project environment.- [PROMPT_INJECTION]: The skill processes upstreamSYSdocuments, creating an indirect prompt injection surface where instructions potentially embedded in the requirements content could influence the agent's behavior during the decomposition process. - Ingestion points: The skill reads system requirements from the
docs/06_SYS/directory and accepts text input via the--promptparameter. - Boundary markers: There are no explicit delimiters defined in the instructions to isolate processed text from the agent's logic.
- Capability inventory: The skill possesses the ability to create and write multiple files in the
docs/07_REQ/directory and execute shell commands for file management and script execution. - Sanitization: Validation is performed for structural compliance and quality scoring, but the skill lacks specific sanitization mechanisms to filter or escape instructions embedded within the requirements data.
Audit Metadata