kshitijthakkar commited on
Commit
a18d50d
Β·
1 Parent(s): c1a84e8

docs: Add open source foundation section to README and app.py

Browse files

Add comprehensive documentation about the foundational open-source projects:
- TraceVerde (genai_otel_instrument) - OTEL instrumentation
- SMOLTRACE - Agent evaluation engine

This provides important context about the complete ecosystem that powers
TraceMind MCP Server and gives proper credit to the underlying projects.

Files changed (2) hide show
  1. README.md +18 -0
  2. app.py +13 -0
README.md CHANGED
@@ -32,6 +32,24 @@ tags:
32
 
33
  TraceMind MCP Server is a Gradio-based MCP (Model Context Protocol) server that provides a complete MCP implementation with:
34
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
35
  ### πŸ› οΈ **7 AI-Powered Tools**
36
  1. **πŸ“Š analyze_leaderboard**: Generate insights from evaluation leaderboard data
37
  2. **πŸ› debug_trace**: Debug specific agent execution traces using OpenTelemetry data
 
32
 
33
  TraceMind MCP Server is a Gradio-based MCP (Model Context Protocol) server that provides a complete MCP implementation with:
34
 
35
+ ### πŸ—οΈ **Built on Open Source Foundation**
36
+
37
+ This MCP server is part of a complete agent evaluation ecosystem built on two foundational open-source projects:
38
+
39
+ **πŸ”­ TraceVerde (genai_otel_instrument)** - Automatic OpenTelemetry Instrumentation
40
+ - **What**: Zero-code OTEL instrumentation for LLM frameworks (LiteLLM, Transformers, LangChain, etc.)
41
+ - **Why**: Captures every LLM call, tool usage, and agent step automatically
42
+ - **Links**: [GitHub](https://github.com/Mandark-droid/genai_otel_instrument) | [PyPI](https://pypi.org/project/genai-otel-instrument)
43
+
44
+ **πŸ“Š SMOLTRACE** - Agent Evaluation Engine
45
+ - **What**: Lightweight, production-ready evaluation framework with OTEL tracing built-in
46
+ - **Why**: Generates structured datasets (leaderboard, results, traces, metrics) that this MCP server analyzes
47
+ - **Links**: [GitHub](https://github.com/Mandark-droid/SMOLTRACE) | [PyPI](https://pypi.org/project/smoltrace/)
48
+
49
+ **The Flow**: `TraceVerde` instruments your agents β†’ `SMOLTRACE` evaluates them β†’ `TraceMind MCP Server` provides AI-powered analysis of the results
50
+
51
+ ---
52
+
53
  ### πŸ› οΈ **7 AI-Powered Tools**
54
  1. **πŸ“Š analyze_leaderboard**: Generate insights from evaluation leaderboard data
55
  2. **πŸ› debug_trace**: Debug specific agent execution traces using OpenTelemetry data
app.py CHANGED
@@ -4,6 +4,19 @@ TraceMind MCP Server - Hugging Face Space Entry Point (Track 1)
4
  This file serves as the entry point for HuggingFace Space deployment.
5
  Exposes 7 AI-powered MCP tools + 3 Resources + 3 Prompts via Gradio's native MCP support.
6
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7
  Architecture:
8
  User β†’ MCP Client (Claude Desktop, Continue, Cline, etc.)
9
  β†’ MCP Endpoint (Gradio SSE)
 
4
  This file serves as the entry point for HuggingFace Space deployment.
5
  Exposes 7 AI-powered MCP tools + 3 Resources + 3 Prompts via Gradio's native MCP support.
6
 
7
+ Built on Open Source Foundation:
8
+ πŸ”­ TraceVerde (genai_otel_instrument) - Automatic OpenTelemetry instrumentation
9
+ for LLM frameworks (LiteLLM, Transformers, LangChain, etc.)
10
+ GitHub: https://github.com/Mandark-droid/genai_otel_instrument
11
+ PyPI: https://pypi.org/project/genai-otel-instrument
12
+
13
+ πŸ“Š SMOLTRACE - Agent evaluation engine with OTEL tracing built-in
14
+ Generates structured datasets (leaderboard, results, traces, metrics)
15
+ GitHub: https://github.com/Mandark-droid/SMOLTRACE
16
+ PyPI: https://pypi.org/project/smoltrace/
17
+
18
+ The Flow: TraceVerde instruments β†’ SMOLTRACE evaluates β†’ TraceMind analyzes
19
+
20
  Architecture:
21
  User β†’ MCP Client (Claude Desktop, Continue, Cline, etc.)
22
  β†’ MCP Endpoint (Gradio SSE)