onboarding

Pass

Audited by Gen Agent Trust Hub on Apr 2, 2026

Risk Level: SAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill contains an optional 'Share to Proof' feature that uploads the generated ONBOARDING.md file to https://www.proofeditor.ai/share/markdown. This results in repository structure and code summaries being sent to an external service.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it processes untrusted data from repository files. Ingestion points: The skill reads README.md, entry points, and source code files during Phase 2. Boundary markers: No delimiters or warnings are used to isolate untrusted file content. Capability inventory: The agent has access to file-reading tools, script execution (node), and network tools (curl). Sanitization: No content validation or sanitization is performed on the ingested data.
  • [COMMAND_EXECUTION]: The skill executes a bundled Node.js script (node scripts/inventory.mjs) to gather metadata and uses shell commands (cat, jq, curl) for optional data sharing.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 2, 2026, 03:11 PM
Security Audit — agent-trust-hub — onboarding