skills/mowgli-ai/skills/mowgli/Gen Agent Trust Hub

mowgli

Pass

Audited by Gen Agent Trust Hub on Jul 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute the mowgli command-line utility via npx to interact with project metadata, version history, and design files.
  • [EXTERNAL_DOWNLOADS]: The skill uses npx mowgli-cli, which downloads the vendor's command-line interface from the NPM registry if it is not already locally available.
  • [DATA_EXFILTRATION]: As part of its intended design synchronization feature, the skill transmits local source code (React components) and configuration files (Tailwind themes) to the remote Mowgli service at app.mowgli.ai.
  • [PROMPT_INJECTION]: The skill implements security best practices by recommending the use of explicit boundary markers (--- BEGIN FILE: <name> --- and --- END FILE: <name> ---) when interpolating untrusted codebase content into instructions for the Mowgli agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 1, 2026, 04:51 PM