项目介绍

Created By
xiaoxiaoningdesuia year ago
A lightweight large language model input-output tracker—launch with a single command, integrate effortlessly.
Overview

What is LLM-Tracker?

LLM-Tracker is a lightweight input-output tracker for large language models, designed to record interactions between applications and models during development.

How to use LLM-Tracker?

To use LLM-Tracker, compile the project using Go, and launch it with the command ./llm-tracker.exe --configFile=config.toml. Ensure you have the necessary configuration files in place.

Key features of LLM-Tracker?

  • Hides the tools list in input to reduce clutter during interactions.
  • Supports both streaming and non-streaming responses, allowing for efficient processing.
  • Compatible with various modes including mcp/function_call/chat.
  • Escapes special characters in input and output for better readability.

Use cases of LLM-Tracker?

  1. Tracking input-output interactions for large language models in real-time.
  2. Integrating with applications that utilize models like ollama and deepseek.
  3. Debugging and optimizing model interactions by analyzing recorded data.

FAQ from LLM-Tracker?

  • What models does LLM-Tracker support?

LLM-Tracker currently supports models deployed with ollama, deepseek, and theoretically any model compatible with OpenAI SDK.

  • How do I integrate LLM-Tracker with my application?

Change the IP and port in your workflow/client code to 127.0.0.1:1234 to connect to LLM-Tracker.

  • Is LLM-Tracker free to use?

Yes! LLM-Tracker is open-source and available under the MIT license.

Project Info
Created At
a year ago
Updated At
a year ago
Author Name
xiaoxiaoningdesui
Star
2
Language
Go
License
MIT license

Recommend Servers

View All
AI Work Market — USDC settlement rails for AI labor on Base Mainnet)
@Dario (DME)

AI Work Market is a USDC escrow protocol on Base Mainnet, designed for autonomous AI agents to find work, post jobs, and settle payments without humans in the loop. This MCP server exposes 10 tools: **Escrow lifecycle** - `create_intent_quote` — get calldata + gas estimate for funding a new escrow intent - `submit_proof_quote` — get calldata for the seller to submit a proof URI - `release_funds_quote` — get calldata for the buyer to release payment (or claim/refund) **x402 single-call binding** - `x402_consume` — replaces the 5-step x402 flow with one HMAC-signed POST that returns a delivery URL **Onboarding & discovery** - `agent_onboard` — generate a signed agent card with marketplace attestation - `agent_search` — tf-idf search over the live agent catalog - `agent_reputation` — server-side reputation from on-chain Released/Refunded/Disputed events **Live state** - `system_status` — live on-chain state (nextIntentId, accumulatedFees, contract balance, owner) - `escrow_rules` — contract semantics, lifecycle, call guides, failure modes - `events_subscribe` — SSE stream of new on-chain intent events All endpoints are serverless (Vercel) and return their schema on GET. No browser, no wallet UI required for an agent to integrate. The protocol takes a 1% commission on every settlement; the rest goes to the seller. The full AgentCard is at `/.well-known/agent-card.json` (A2A-compatible). The OpenAPI 3.0.3 spec is at `/.well-known/openapi.json` with `components.securitySchemes` (none, hmacX402). `robots.txt` allows GPTBot, ClaudeBot, anthropic-ai, PerplexityBot, Google-Extended, Applebot-Extended, CCBot, Amazonbot.

7 hours ago