pickfit

BrowseDeveloper CLIs

Ducklab

Self-hosted harness that runs full dev cycle with multiple LLMs under test gates.

Tested on LinuxKnown vulnerabilities

Use it

Before you install: known vulnerabilities

pickfit found 1 known high or critical vulnerability in the packages a fresh install brings (OSV.dev, 3 Nov 2026). Weigh this before you install it.

  • go:golang.org/x/image v0.40.0 golang.org/x/image/tiff has excessive resource consumption in PackBits decompression (GHSA-q675-qj96-32m9, high, fixed in 0.41.0)

Linux x86_64

You need first

  • Runtimes: Go 1.25+, Node 22+
  • System packages: git, libgtk-3-dev libwebkit2gtk-4.1-dev

These steps ran from a folder named work in the home folder, which already existed. Make it and run them from there:

mkdir -p "$HOME/work" && cd "$HOME/work"
  1. git clone the ducklab repo and go build the ducklab CLI, ducklab-engine and ducklab binaries into ~/.local/bin (candidate 3; candidates 1 and 2 skipped: apt needs root, GTK dev headers absent)

    cd "$HOME"/work && git clone --depth 1 https://github.com/jrullan/ducklab && cd ducklab && mkdir -p ~/.local/bin && go build -o ~/.local/bin/ducklab-cli ./cmd/ducklab && go build -o ~/.local/bin/ducklab ./cmd/ducklab && go build -o ~/.local/bin/ducklab-engine ./cmd/ducklab-engine
Where the README and reality differ
  • README: go build -o ducklab-cli ./cmd/ducklab (then ducklab commands used as on PATH)
    What worked: The CLI auto-starts the engine by looking for ducklab-engine on PATH, so ducklab-engine must be built and installed too (make install does this). Built all three binaries into ~/.local/bin.
  • README: ducklab duckling scorecard
    What worked: No such verb: 'unknown duckling command: scorecard'. Duckling verbs are list, add, probe, test, remove, show, set, suggest. Seat suggestions with pass rates come from 'ducklab roster suggest'.
Verified in pickfit: the commands exactly as its sandbox ran them

For the record only: they name the sandbox's home, /home/sbx.

cd /home/sbx/work && git clone --depth 1 https://github.com/jrullan/ducklab && cd ducklab && mkdir -p ~/.local/bin && go build -o ~/.local/bin/ducklab-cli ./cmd/ducklab && go build -o ~/.local/bin/ducklab ./cmd/ducklab && go build -o ~/.local/bin/ducklab-engine ./cmd/ducklab-engine
ducklab --version
cat > /home/sbx/work/mcp_init.jsonl <<'EOF'
{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"pickfit-check","version":"0.0.1"}}}
{"jsonrpc":"2.0","method":"notifications/initialized"}
{"jsonrpc":"2.0","id":2,"method":"tools/list","params":{}}
EOF
ducklab mcp serve < /home/sbx/work/mcp_init.jsonl

Check that it works

ducklab --version

The output should contain “ducklab dev”.

First thing to try

cat > "$HOME"/work/mcp_init.jsonl <<'EOF'
{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"pickfit-check","version":"0.0.1"}}}
{"jsonrpc":"2.0","method":"notifications/initialized"}
{"jsonrpc":"2.0","id":2,"method":"tools/list","params":{}}
EOF
ducklab mcp serve < "$HOME"/work/mcp_init.jsonl

What it does

  • Generate requirements, spec and plan from a brief
    Not tested
  • Run tasks in solo, pair, tournament, split, council modes
    Not tested
  • Assign model seats from measured scorecards
    Not shown working
  • Operate the loop from another model via MCP
    Not shown working

How far pickfit got

Linux x86_64

  1. Installs
  2. Starts
  3. 0 of 4 features work
  4. Works offline

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

Good for

  • write requirements, spec and plan from a brief
  • build development tasks with one or several models
  • run project test gate for verification
  • review task commit
  • plan release notes of what shipped

How pickfit tested it

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

Linux x86_64

Its main function works

  • Replayed from scratch
  • Works offline
  • Checked on 16 Nov 2026
  • Version dev

Differences the test machine hit

  • README: sudo apt install libgtk-3-dev libwebkit2gtk-4.1-dev; make desktop && make install
    In the test: Desktop needs root for apt and GTK/WebKit dev headers (pkg-config gtk+-3.0 missing on this box). Not attempted; only the CLI and engine were built.
  • README: ducklab roster suggest / cycle commands run from a project
    In the test: Commands need a ducklab project: 'no ducklab project at ... or any parent' outside one. Create it with git init and ducklab project init --name ...
  • README: Needs Go 1.25+
    In the test: Built with the box's Go 1.27.4 with no problems; the module downloads needed network during the build step.

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.