Jonathan Martín
@jonimartin27
Autonomous software studio (Fervon): AI agent fleets building local-first tooling — Lookspan, inferbench, Trace, Veredicto. Also software built to order.
7
Published Tools
2
Total Stars
100
Weekly Downloads
1
GitHub Followers
24
Public Repos
Published Tools
4 MCP Servers2 Skills1 Agentacross 4 categoriesJoniMartin27
Benchmark local LLM inference speed (tokens/sec) on your own hardware — llama.cpp native + cloud APIs, 124-model catalog, optimal-quant picker, and an MCP serve mode.
...moreJonathan Martin
Lookspan adapter for CrewAI — emits spans for crew kickoffs, agent execution, tasks, tools, and LLM calls.
Jonathan Martin
Lookspan adapter for LangChain / LangGraph — emits spans for chains, LLM calls, and tools.
lookspan
Lookspan SDK for MCP — wraps any MCP client to emit spans for each tool call.
lookspan
Drop-in tracing for the OpenAI SDK — wrap your client in one line and every model call becomes a Lookspan span.
lookspan
Drop-in tracing for the Anthropic SDK — wrap your client in one line and every message call becomes a Lookspan span.
Jonathan Martin
Local-first observability for AI agents: emit spans from LangGraph, CrewAI, MCP or your own code to a dashboard that runs on your machine — no cloud, no account, no vendor.
...more