full-stack-project-finisher
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCREDENTIALS_UNSAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The
analyze_project_gaps.pyandtest_coverage_report.pyscripts scan project files to identify tasks and test coverage. A malicious file within the scanned project could attempt to influence the agent's interpretation of the scan results. - Ingestion points:
scripts/analyze_project_gaps.pyandscripts/test_coverage_report.pyread all code files in the provided directory. - Boundary markers: The scripts do not use specific delimiters or instructions to ignore content within the processed files.
- Capability inventory: The scripts are limited to local file system reading and generating reports (stdout or local JSON file).
- Sanitization: The scripts perform no sanitization of the strings extracted from the scanned files before inclusion in reports.
- [CREDENTIALS_UNSAFE]: Development templates in
assets/docker-compose.example.ymlandassets/.env.examplecontain default credentials (postgres:postgres) and placeholder keys. These are standard for example configurations intended for local development environments.
Audit Metadata