mechanical-teardown
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The core functionality is implemented in
scripts/teardown.py, which processes teardown data to generate engineering metrics and reports. The script relies entirely on standard Python libraries and does not perform network operations or access sensitive system locations. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from external teardown guides and user-provided JSON specifications. This represents a potential vulnerability surface; however, the script uses strict JSON parsing and mandatory HTML escaping (
html.escape) for all interpolated text, preventing common injection attacks or malicious code rendering in the output artifacts. - [COMMAND_EXECUTION]: Automated analysis flagged the use of
subprocess.runintests/test_teardown.py. Investigation shows these calls are used solely to execute the skill's own CLI tools during the test suite, which is a benign and standard development practice. - [DATA_EXPOSURE]: The report generation logic allows embedding local images as base64 data. While this allows reading local files, it is an intended feature for including teardown photos in the self-contained HTML report and does not involve automated exfiltration to remote servers.
Audit Metadata