skills/udecode/better-convex/dig-dig/Gen Agent Trust Hub

dig-dig

Pass

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: Uses git clone and ls to retrieve and verify repository content for documentation research. This allows the agent to access the latest source code of libraries.- [EXTERNAL_DOWNLOADS]: Fetches library source code from GitHub repositories to provide up-to-date documentation and API information.- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing untrusted content from third-party repositories.
  • Ingestion points: Files downloaded from external GitHub repositories are processed by a subagent.
  • Boundary markers: The instructions do not specify delimiters to separate untrusted code from the agent's system instructions.
  • Capability inventory: Includes shell command execution and subagent tasking for file exploration.
  • Sanitization: The skill does not perform content filtering or sanitization on the source code before exploration.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 01:20 AM