find-untested-sources

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and parses source files from the provided repository root to identify declared symbols and import relationships.
  • Ingestion points: Repository source files are read by scripts/Find-UntestedSources.cs and scripts/find_untested_sources.py.
  • Boundary markers: Not present; extracted identifiers and file paths are included directly in the JSON output report.
  • Capability inventory: The scripts perform read-only analysis and print JSON to standard output; no network, filesystem write, or command execution capabilities are present.
  • Sanitization: None; the tool reports code identifiers exactly as they appear in the source text.
  • [EXTERNAL_DOWNLOADS]: The skill requires well-known, established libraries for code analysis, specifically the tree-sitter-language-pack Python package and the Microsoft.CodeAnalysis.CSharp NuGet package.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 07:35 AM
Security Audit — agent-trust-hub — find-untested-sources