seed-hypermedia

Warn

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS] (MEDIUM): The skill utilizes 'npx -y' to download and execute the '@seed-hypermedia/cli' package from the npm registry. The publisher is not within the trusted organization scope.
  • [REMOTE_CODE_EXECUTION] (MEDIUM): The use of 'npx -y' bypasses the user confirmation step, allowing for the immediate execution of remote code on the host system.
  • [CREDENTIALS_UNSAFE] (LOW): The CLI tool includes subcommands for 'key management' such as 'import', 'show', and 'generate' for cryptographic mnemonics and signing keys, which are sensitive credentials.
  • [COMMAND_EXECUTION] (LOW): The skill employs shell commands including 'curl' and 'npx' to perform its operations.
  • [PROMPT_INJECTION] (LOW): The skill processes remote hypermedia content which could contain malicious instructions. Evidence Chain: 1. Ingestion points: 'get ' command fetches remote content from 'hm://' URLs. 2. Boundary markers: None identified. 3. Capability inventory: 'npx' execution, 'curl' operations, and cryptographic key management. 4. Sanitization: No sanitization logic described.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 17, 2026, 06:20 PM