Fsi

What it does

When your agent uses it

• R nuget references
• Load composition
• Interactive type exploration
• Small strongly typed experiments before moving code into a project

Example prompts

• “/fsi”

Requirements

• Compatibility (from SKILL.md): Requires a .NET SDK with `dotnet fsi`; NuGet-backed scripts may need package restore and trusted package sources.

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

1. 1Decide whether the request is a disposable REPL probe, a repeatable .fsx script, or code that should be promoted to a compiled F# project.
2. 2Put repeatable work in an .fsx file immediately. Add all required #r, #load, open, input path, and package source directives to the script…
3. 3Keep package references pinned when the script should be reproducible, and use only trusted NuGet feeds or local feeds derived from…
4. 4Run the script with dotnet fsi from a clean shell, passing the same arguments the user or CI will use.
5. 5Promote the script to an .fsproj when it needs tests, distribution, project references, or long-term CI coverage.

What it can do on your machine

Read from SKILL.md and the folder at commit 535dd55. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • dotnet

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  Links to these hosts (documentation or services it may open):

  • learn.microsoft.com

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

• Compatibility

  Requires a .NET SDK with `dotnet fsi`; NuGet-backed scripts may need package restore and trusted package sources.

  From compatibility in the SKILL.md frontmatter.

Context cost

Fsi loads about 1.7k tokens when it runs. Until then it costs about 163 tokens; SKILL.md has 570 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~163

When it runs · the whole SKILL.md, loaded when a task matches

~1.7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from managedcode/dotnet-skills at commit 535dd55, republished under its MIT licence (© managedcode). 570 words, ~1,663 tokens.

name

fsi

description

Use F# Interactive (`dotnet fsi`) for .NET exploration, scriptable experiments, package-backed .fsx workflows, quick data transforms, and reproducible command-line probes. USE FOR: .fsx scripts, F# REPL work, #r nuget references, #load composition, interactive type exploration, and small strongly typed experiments before moving code into a project. DO NOT USE FOR: production application code that needs compiled project structure; C# scripting; long-lived automation better expressed as a normal CLI, test, or build target. INVOKES: run dotnet fsi, edit .fsx scripts, load project or source files, and validate snippets against the target SDK.

compatibility

Requires a .NET SDK with `dotnet fsi`; NuGet-backed scripts may need package restore and trusted package sources.

dotnet fsi

dotnet fsi scripts/check.fsx

#!/usr/bin/env -S dotnet fsi

printfn "Hello from FSI"

chmod +x scripts/check.fsx
./scripts/check.fsx

let square x = x * x;;

[ 1 .. 5 ] |> List.map square;;

open System
open System.IO

let summarize path =
    File.ReadLines path
    |> Seq.filter (String.IsNullOrWhiteSpace >> not)
    |> Seq.countBy (fun line -> line.Split(' ')[0])
    |> Seq.sortByDescending snd
    |> Seq.truncate 10
    |> Seq.toList

for key, count in summarize "input.log" do
    printfn $"{key}: {count}"

open System
open System.IO

let input = "data/items.txt"
let output = "artifacts/items.normalized.txt"

if not (File.Exists input) then
    failwith $"Missing input file: {input}"

Directory.CreateDirectory(Path.GetDirectoryName output) |> ignore

File.ReadLines input
|> Seq.map (fun line -> line.Trim())
|> Seq.filter (String.IsNullOrWhiteSpace >> not)
|> Seq.distinct
|> Seq.sort
|> fun lines -> File.WriteAllLines(output, lines)

#r "nuget: Newtonsoft.Json, 13.0.3"

open Newtonsoft.Json

let payload = {| Name = "Ada"; Kind = "sample" |}
let json = JsonConvert.SerializeObject(payload)

printfn $"{json}"

#r "nuget: FSharp.Data, 6.6.0, usepackagetargets=true"

let localFeed =
    System.IO.Path.Combine(__SOURCE_DIRECTORY__, "../artifacts/packages")
    |> System.IO.Path.GetFullPath

#i $"nuget: {localFeed}"

// MathHelpers.fsx
let square x = x * x

// Check.fsx
#load "MathHelpers.fsx"
open MathHelpers

printfn $"%d{square 12}"

dotnet new console -lang "F#" -o tools/Probe
dotnet build tools/Probe/Probe.fsproj

dotnet fsi scripts/check.fsx
dotnet fsi scripts/check.fsx -- arg1 arg2

© managedcode, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in catalog/Tools/FSharp/skills/fsi of managedcode/dotnet-skills.

• SKILL.md
• manifest.json

Open the folder on GitHubat commit 535dd55