flow-setup
Skill1 starsChanged 3 months agoMITView on GitHub
Install the Flow programming language, verify the toolchain, and fix a broken Flow environment. Use when Flow is not installed, when `flow` is not found, when a first program fails to build, when the C compiler or Python host is missing, or when the user is starting from nothing.
- Installs packages
2 other projects use a version of this.
Adopted by
Most recently taken up by
- cofin/flowtheir own version, first seen Nov 7, 2026
- Indrajitvp/claude-skillstheir own version, first seen Oct 1, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- cofin/flow★ 19same name
- Indrajitvp/claude-skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository13 other files it gives its agents.
- AGENTS.mdAGENTS.md
- flow-audioSkill
- flow-autodiffSkill
- flow-basicsSkill
- flow-compiler-devSkill
- flow-debugSkill
- flow-dynamicsSkill
- flow-effectsSkill
- flow-graphicsSkill
- flow-perfSkill
- flow-stdlibSkill
- flow-verifySkill
- flow-wasmSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
Run the packaged tool first.
- Very specific: line 18
If flow-doctor is not on the path, do the checks by hand:
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Setting up Flow
Get from a bare machine to a program that runs, and prove it.
Check what is already there
Run the packaged tool first. It checks everything below in one pass.
flow-doctorIf flow-doctor is not on the path, do the checks by hand:
flow version || ./flow version # is Flow installed
cc --version # is there a C compiler
python3 --version # is there a Python host
echo "FLOW_HOST=${FLOW_HOST:-unset}"Install Flow
macOS, with Homebrew:
brew tap flooooooooooow/flow
brew install flow
flow versionFrom source, which also gets the examples and tests:
git clone https://github.com/flooooooooooow/flow.git
cd flow
./flow versionFrom source the command is ./flow, run from inside the checkout. Set
FLOW_HOME to that directory so the tools in this pack can find the standard
library and examples:
export FLOW_HOME="$HOME/flow"Install the build tools
Flow compiles to C and then to a native binary, so a C compiler is required.
macOS:
xcode-select --installDebian or Ubuntu:
sudo apt install build-essential python3Set the compiler host
This is the step people miss.
| Host | Command | Covers |
flowc (default) | flow run f.flow | Functions, structs, fixed arrays, pointers, loops, basic match, extern |
python | FLOW_HOST=python flow run f.flow | Everything, including println, effects, flow blocks, generics, MLIR, GPU |
println is outside the Stage-A subset, so almost every beginner program needs
the Python host. Recommend setting it permanently:
echo 'export FLOW_HOST=python' >> ~/.zshrc # or ~/.bashrc
source ~/.zshrcProve it works
cat > /tmp/flow_smoke.flow <<'EOF'
function main() -> i32 {
println("Hello, Flow!")
return 0
}
EOF
FLOW_HOST=python flow run /tmp/flow_smoke.flowExpected tail:
Hello, Flow!
✅ Program finished
Exit code: 0The first build takes roughly twenty seconds. That is the C toolchain doing real work, not a hang.
Optional pieces
| Want | Install |
| MLIR backend, JIT | flow setup installs LLVM tooling; then --backend=mlir |
| Graphics on Linux or Windows | SDL2 development packages |
| Editor support | flow lsp, plus the VS Code extension from the main repository |
| A project rather than a loose file | flow init myproject, then flow build |
Diagnosis table
| Symptom | Cause | Fix |
command not found: flow | Not installed, or source checkout | Install with Homebrew, or use ./flow inside the checkout |
flowc emit failed (Stage-A subset?) | Program needs the full language | Prefix FLOW_HOST=python |
clang: command not found | No C compiler | Install the build tools above |
ModuleNotFoundError on the Python host | Python dependencies missing | pip3 install -r requirements.txt inside the checkout |
| Builds land somewhere unexpected | Output goes to $FLOW_HOME/build/ | Look there, and use --out where a command supports it |
Report back
Give the user the exact command that worked, its real output, and the host that was used. If you changed a shell profile, say which file and which line.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

