self-improve-with-tinyfish

Pass

Audited by Gen Agent Trust Hub on May 13, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses shell commands to check for the presence of the TinyFish CLI and its authentication status (which tinyfish, tinyfish auth status). This is part of its core functionality for verifying required dependencies before proceeding.
  • [COMMAND_EXECUTION]: The skill executes Python scripts as fallbacks for web searching and content fetching when the CLI is unavailable. These scripts interact with the vendor's official APIs (api.search.tinyfish.ai and api.fetch.tinyfish.ai) and handle user-defined topics securely via environment variables.
  • [EXTERNAL_DOWNLOADS]: The skill is configured to be installed from the author's official GitHub repository (github.com/tinyfish-io/tinyfish-cookbook), which is consistent with the vendor's identity.
  • [DATA_EXFILTRATION]: While the skill performs network operations to search and fetch web content, it exclusively targets the vendor's official API domains. It does not access sensitive local files or attempt to exfiltrate user credentials.
Audit Metadata
Risk Level
SAFE
Analyzed
May 13, 2026, 04:36 AM