kotlin-multiplatform-skill-harvester

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Executes a bundled Python script scripts/harvest_lessons.py to process local project files and generate reports.
  • [COMMAND_EXECUTION]: Uses the GitHub CLI (gh) to interact with issues on the author's repository (ronjunevaldoz/kmm-agent-skills) for fetching consumer feedback.
  • [EXTERNAL_DOWNLOADS]: Retrieves external data from GitHub issues via the gh command-line tool.
  • [PROMPT_INJECTION]: Processes untrusted consumer-provided content from docs/lessons/*.md files, which acts as an indirect prompt injection surface.
  • Ingestion points: Files located in docs/lessons/*.md across various consumer projects.
  • Boundary markers: Content is parsed based on markdown headings and YAML frontmatter structure.
  • Capability inventory: Execution of gh CLI commands and file system modifications (applying diffs to other skills).
  • Sanitization: The skill explicitly requires a manual review phase (Step 5) and user confirmation before any modifications are committed to the file system.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 03:15 AM
Security Audit — agent-trust-hub — kotlin-multiplatform-skill-harvester