DiscoverMakersCommunitySubmit tool
Submit project
WhatWeMade

Open-source and AI-assisted projects, and exactly how people made them. Free for makers, forever.

Or choose topics. Unsubscribe any time.

Buy me a coffee

Explore

  • Projects
  • Makers
  • Websites

Community

  • Discussions
  • Build week
  • Guidelines
  • Newsletter

For makers

  • Submit a project
  • Publish with your agent
  • Asset pack prompt
  • Build recipe prompt
  • Template prompt
  • whatwemade.json
  • Marketplace

About

  • How it works
  • How we review
  • How ranking works
  • Contact
© 2026 whatwemade.appPrivacyCookiesTermsTakedownsMade with Claude Code

Explore

Tools/Engineering

Llmxray

LLMxRay — Local LLM Observatory. Full observability interface for Ollama: chat, streaming, reasoning, RAG, introspection, metrics.

DjovaniLaunched 7 days agoWeb app, MCP serverRuns on desktopWorking · checked 3 hours ago
1 / 4
View on GitHubTry it live

About

See what your local AI is actually doing.

Point VS Code Copilot, Open WebUI, Continue, n8n or your own agent at LLMxRay instead of Ollama, and see every turn: what was cached, what was read again, what the daemon silently cut, which tool evicted which model. No cloud. No API keys. No cost.

You need Ollama running with at least one model pulled (ollama pull llama3.2). Then:

#ai-agents#developer-tools#llm-observability#local-llm#logprobs

How it was made

Estimated by the maker.

More like this

Agents

  • Claude Code

    100% of the work

  • Claude · inside the product

  • ChatGPT · inside the product

  • Codex · inside the product

  • Cursor · inside the product

  • Gemini · inside the product

  • Gk

    Grok · inside the product

  • Copilot · inside the product

Stack

MCP SDKTailwindTypeScript

MCP servers

No MCP servers. Built with the agents alone.

Discussion

0 comments

Sign in to join in. A kind word or a good question goes a long way.

    View on GitHubTry it live

    What you need

    Quick to run

    Tools

    • Node 22+

    LogneBudo/llmxray

    28

    stars

    0

    versions here

    Apache-2.0

    license

    Last commit 1 week ago.

    Types
    Web app, MCP server
    Category
    Engineering
    Availability
    Run it yourself
    Live at
    lognebudo.github.io/llmxray

    Made by

    • Djovani

      Solution Architect exploring many topics

    awesomeopen.github.io
    Awesome Open* Catalog· Web app

    A catalog of software named Open*, with WebMCP tools for AI agents

    Owlie· Web app

    Monitor feeds, videos and podcasts and get structured results

    Olivares AI· Web app

    Run AI coding agents on your own servers, with a signed audit trail

    AIL Context Compressor — zsh — 120×30

    $ A=/tmp/ailx/ail-context-compressor/scripts/ail_mcp.py

    $ python3 "$A" pack -f demo-notes.txt --next "review title and copy with Captain" | tee demo-snapshot.ail

    --- tokens: 49 -> 66 (-34.7% saved, 0 @SIG, redactions 0) ---

    !AILv1

    @CTX DEMO DATA ONLY — fictional Captain WhatWeMade draft handover.

    @CTX Working title: Harborlight weekend market.

    @CTX Status: outline only; nothing has been published.

    @CTX Next: review title and copy with Captain.

    @NEXT ?review_title_and_copy_with_Captain

    $ python3 "$A" unpack -f demo-snapshot.ail

    DEMO DATA ONLY — fictional Captain WhatWeMade draft handover.

    Working title: Harborlight weekend market.

    Status: outline only; nothing has been published.

    Next: review title and copy with Captain.

    $

    AIL Context Compressor· MCP server

    Portable context handovers between AI chats and agents