pickfit

BrowseLocal AI tools

Mithril

Orchestration engine that routes requests across LLM providers via one Ollama-compatible API.

Tested on macOS and LinuxKnown vulnerabilities

Use it

Before you install: known vulnerabilities

pickfit found 2 known high or critical vulnerabilities in the packages a fresh install brings (OSV.dev, 3 Nov 2026). Weigh this before you install it.

  • crates:quinn-proto 0.11.17 Peer minimum ACK delay can panic quinn-proto when ACK_FREQUENCY transmission is enabled (RUSTSEC-2026-0340, high, fixed in 0.11.18)
See 1 more

macOS arm64

You need first

  • Keys or accounts: gemini, openai
  1. brew install Mithril via GiacomoSaccaggi/tap

    brew install GiacomoSaccaggi/tap/mithril 2>&1
    

Check that it works

mithril --version

The output should contain “mithril 0.3.0”.

First thing to try

mithril fellowship init
ls -R .mithril 2>&1
cat .mithril/fellowship.yaml 2>&1 | head -n 200
mithril serve --port 16180
curl -s http://localhost:16180/api/tags

Run mithril serve --port 16180 to start the server, then point your client at http://localhost:16180.

Linux x86_64

You need first

  • Keys or accounts: gemini, openai
  1. Install mithril-cli via uv tool

    uv tool install mithril-cli

Check that it works

mithril --version 2>&1

The output should contain “mithril 1.0.0”.

First thing to try

mithril fellowship init 2>&1; echo "---EXIT:$?"; ls -R .mithril 2>&1; cat .mithril/fellowship.yaml 2>&1
mithril serve
curl -s http://localhost:16180/api/tags 2>&1

Run mithril serve --port 16180 to start the server, then point your client at http://localhost:16180.

What it does

  • Combine LLM providers behind one Ollama-compatible API
    How to use it
    1. Run mithril fellowship init to create a fellowship config in .mithril/fellowship.yaml.
    2. Start the server with mithril serve --port 16180.
    3. List the available models with curl -s http://localhost:16180/api/tags.
    Tested
  • Use 24 built-in file, git and web tools
    Tested
  • Serve Ollama, OpenAI and MCP API endpoints
    How to use it
    1. Start the server with mithril serve --port 16180 and check it with curl -s http://localhost:16180/health.
    2. List models on the OpenAI-style endpoint with curl -s http://localhost:16180/v1/models.
    3. List MCP tools with curl -s -X POST http://localhost:16180/mcp -H "Content-Type: application/json" -d '{"jsonrpc":"2.0","id":1,"method":"tools/list","params":{}}'.
    Tested
  • Run a team of AI models as a fellowship
    Not tested

How far pickfit got

macOS arm64

  1. Installs
  2. Starts
  3. 3 of 4 features work
  4. Works offline

Linux x86_64

  1. Installs
  2. Starts
  3. 3 of 4 features work
  4. Works offline

Windows: the project says it runs there; pickfit hasn’t tried it.

Good for

  • Backend for Junie
  • Backend for OpenCode
  • Backend for Open WebUI
  • Backend for LangChain / LlamaIndex
  • MCP server for Claude Desktop

How pickfit tested it

Each run starts from a clean machine and follows the steps above. How pickfit verifies

macOS arm64

Its main function works

  • Replayed from scratch
  • Works offline
  • Checked on 1 Nov 2026
  • Version 0.3.0

Linux x86_64

Its service answers requests

  • Replayed from scratch
  • Works offline
  • Checked on 30 Oct 2026
  • Version 1.0.0

Differences the test machine hit

  • README: mithril serve starts the engine on http://localhost:16180
    In the test: server starts without a local model and still serves HTTP; log warns No local model found but health and tags work

How to read a tool card

Stitching
pickfit installed this tool itself. Everything under the stitches comes from that run. A tool pickfit has not run never gets stitches.
Verified on your OS
The install steps ran on the system you chose. “Ran on Linux only” means it has not been run on yours yet.
macOS, Linux and WindowsWhere it ran
Green OS marks: pickfit ran the install steps there. Grey ones: the README says it runs there, and pickfit has not tried.
How far the run got
Installs and starts, its main function works, its service answers, it produces model output. A window means a desktop app opened; a plug, an MCP client used it. The mark sits beside the run’s result, under “Why it fits and how we know” and in “How pickfit tested it”.
Visual check
pickfit looked at screenshots of the app. Green: it looks right. Amber with a number: things that look wrong; open it to see them. Grey and dashed: not checked yet.
Known vulnerabilities
Known high or critical vulnerabilities (OSV.dev) in what it installs. Green: none known. Red with a number: how many; open it for the list, and read it before you install. Grey and dashed: not checked, which does not mean safe.
About the run
The day it ran, and the version that ran.
Source of a reason
Every reason a pick fits rests on one profile field. Open the reason to see the field and the value it relies on.
Hidden gem
High quality, little known.
Quality and popularity
Two scores out of 100. Quality: how well it installed and worked when pickfit tried it, and how good its install docs and upkeep are. Popularity: from stars, downloads and Hacker News.
Features checked
Of the main features the README advertises, how many the run actually exercised (green), and how many failed (red). Grey ones were not checked, and the card says why.
Not verified
A well-known tool Fitz mentions from general knowledge. No run record, no steps, no score.