hld-writer
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local Python scripts (
plugins/testany-eng/scripts/trace_lint.pyandtrace_build_rtm.py) to validate and build traceability metadata for the generated designs. These scripts are provided by the vendor (testany-io/testany-eng) as part of the design workflow. - [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes external, potentially untrusted project documentation (PRD, API Contracts) to inform its output.
- Ingestion points: Reads files found through directory scanning (e.g.,
**/PRD*,**/*contract*). - Boundary markers: The skill defines strict content boundaries for the output HLD, but lacks specific delimiters or sandboxing instructions for the input data parsed from project files.
- Capability inventory: Can execute shell commands via the
python3runner for the vendor-supplied linting tools. - Sanitization: No specific sanitization or filtering of input document content is described.
Audit Metadata