Can this skill help reduce my OpenAI/Anthropic API costs?

Yes, by using the performance analysis workflow, the skill identifies high token usage and inefficient tool calls, helping you optimize your agent's prompt efficiency.

How do I install the necessary tools?

You need to install the langsmith-fetch CLI via pip (pip install langsmith-fetch) and ensure your LangSmith environment variables are exported in your terminal.

What is the LangSmith Fetch skill for Claude?

It is a specialized capability that allows Claude to use the langsmith-fetch CLI to retrieve, analyze, and troubleshoot execution traces from LangChain and LangGraph agents directly in your workspace.

Do I need a LangSmith account to use this skill?

Yes, you must have an active LangSmith account, a project set up, and your LANGSMITH_API_KEY configured in your environment variables.

Can I export my debugging sessions for my team?

Absolutely. The skill includes a specific workflow to export traces and conversation threads into organized local directories for sharing or archival.

LangSmith Fetch Debugger

Name: LangSmith Fetch Debugger
Author: EmaBilibili

byEmaBilibili

0•

分析と監視

Debugs LangChain and LangGraph agents by fetching and analyzing execution traces directly from LangSmith Studio within the terminal.

LangSmith Fetch is a specialized debugging skill designed for developers building complex AI agents with LangChain and LangGraph. It bridges the gap between local development and LangSmith Studio by allowing Claude to retrieve execution traces, analyze tool call sequences, and investigate memory operations in real-time. Whether you are troubleshooting agent failures, identifying performance bottlenecks, or auditing token usage, this skill provides automated CLI workflows to pinpoint root causes and suggest fixes without leaving your terminal environment.

主な機能

010 GitHub stars

02Memory operation inspection to verify Long-Term Memory (LTM) consistency

03Deep-dive diagnostics for tool calls, results, and failure points

04Performance auditing for execution latency and token usage

05Automated trace retrieval and multi-level execution analysis

06Session export capabilities for archiving and team collaboration

ユースケース

01Troubleshooting 'forgetful' agents by auditing memory store and recall operations

02Optimizing agent response times by identifying slow execution steps and high-cost calls

03Investigating agent logic failures and database connection timeouts in LangGraph

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add emabilibili/my-portfolio langsmith-fetch

For use in Claude.ai and ChatGPT

主な機能

010 GitHub stars

02Memory operation inspection to verify Long-Term Memory (LTM) consistency

03Deep-dive diagnostics for tool calls, results, and failure points

04Performance auditing for execution latency and token usage

05Automated trace retrieval and multi-level execution analysis

06Session export capabilities for archiving and team collaboration

ユースケース

01Troubleshooting 'forgetful' agents by auditing memory store and recall operations

02Optimizing agent response times by identifying slow execution steps and high-cost calls

03Investigating agent logic failures and database connection timeouts in LangGraph

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add emabilibili/my-portfolio langsmith-fetch

For use in Claude.ai and ChatGPT