skf-analyze-source

Pass

Audited by Gen Agent Trust Hub on May 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses subprocesses to execute ast-grep for structural code analysis and the gh command-line tool for interacting with GitHub. These commands are used to identify code patterns, exports, and repository metadata.\n- [EXTERNAL_DOWNLOADS]: The skill performs network operations to access remote repositories on GitHub using the gh API to retrieve release information and file contents during version detection.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) as it ingests and analyzes untrusted external source code, manifest files, and documentation.\n
  • Ingestion points: step-02-scan-project.md and step-04-map-and-detect.md read file structures, manifests, and source code content from the target project.\n
  • Boundary markers: No explicit instruction-ignoring delimiters are mentioned for the source code ingestion phase.\n
  • Capability inventory: The skill possesses subprocess execution (ast-grep, gh), file writing (report and brief generation), and network access capabilities.\n
  • Sanitization: No specific sanitization or filtering logic is described for the analyzed code content.
Audit Metadata
Risk Level
SAFE
Analyzed
May 6, 2026, 03:13 AM