project-migrate

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted project data to infer technology stacks and plan migrations.
  • Ingestion points: The skill reads local project manifest and configuration files (e.g., package.json, requirements.txt, pyproject.toml) to deduce versions and dependencies, as detailed in references/flow.md.
  • Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore or isolate instructions that might be embedded within these external files.
  • Capability inventory: The skill has the capability to write multiple files to the local directory and use web browsing tools (Chrome MCP) to access external content.
  • Sanitization: No sanitization, escaping, or validation of the content read from the source project files is performed before processing.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes a web browsing tool (Chrome MCP) to visit user-provided URLs for the purpose of capturing 'Golden Master' testing data from a legacy environment, as described in references/flow.md and references/golden-master-testing.md. This involves network operations to potentially non-whitelisted or untrusted domains.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 05:04 PM
Security Audit — agent-trust-hub — project-migrate