superthread

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to install a CLI tool from the author's repository ('brew install steveclarke/tap/superthread'). This is a standard vendor-provided dependency.
  • [COMMAND_EXECUTION]: The skill functions by executing the 'suth' command with various subcommands to manage resources on the Superthread platform.
  • [CREDENTIALS_UNSAFE]: The 'suth setup' process involves handling a Superthread API key for authentication, which is an expected behavior for an API-based CLI tool.
  • [PROMPT_INJECTION]: The skill processes content from an external source (Superthread), presenting an indirect prompt injection surface (Category 8).
  • Ingestion points: Data is retrieved via commands like 'suth cards get' and 'suth search query' (SKILL.md).
  • Boundary markers: The skill does not define specific markers to isolate external data from agent instructions.
  • Capability inventory: The agent has the ability to create, update, and delete cards, spaces, and other project items.
  • Sanitization: Content fetched from the platform is processed as provided without additional sanitization.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 10:57 AM