boost-prompt

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides Clojure code meant for execution through the Joyride VS Code extension. The snippet (vscode/env.clipboard.writeText ...) uses the standard VS Code API for clipboard management, which aligns with the skill's stated goal of making refined prompts easily accessible to the user.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it processes user responses to generate a final prompt.
  • Ingestion points: User dialogue and clarification responses entering the refinement loop.
  • Boundary markers: The instructions do not define specific delimiters or instructions to prevent the agent from obeying instructions embedded in user-supplied data.
  • Capability inventory: The skill is limited to the joyride_request_human_input tool and clipboard write operations; it lacks network or file system access permissions.
  • Sanitization: There is no evidence of validation or sanitization for the data ingested from users.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 12:22 PM
Security Audit — agent-trust-hub — boost-prompt