langsmith

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: Fetches and executes an installation script for the LangSmith CLI from the official LangChain repository.
  • [COMMAND_EXECUTION]: Modifies shell profile files (~/.bashrc and ~/.zshrc) to persist required environment variables across sessions.
  • [EXTERNAL_DOWNLOADS]: Installs the LangSmith SDK and related dependencies (openai, openevals) from official package registries.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing external trace and feedback data.
  • Ingestion points: Reads and displays application traces and user feedback from the LangSmith platform (SKILL.md, scripts/quickstart.py).
  • Boundary markers: No explicit boundary markers or instructions to ignore embedded content are defined in the scripts for handling trace data.
  • Capability inventory: The skill uses the Bash and WebFetch tools for system and network operations.
  • Sanitization: Data handling relies on the standard implementation of the LangSmith SDK.
Recommendations
  • INFO: Downloads code from trusted source(s): https://raw.githubusercontent.com/langchain-ai/langsmith-cli/main/scripts/install.sh
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 02:29 AM