dotnet-fsharp: Adding Basic F# Skills#753
Conversation
ycastorium
requested review from
JanKrivanek,
ViktorHofer,
dbreshears and
timheuer
as code owners
|
Note This PR is from a fork and modifies infrastructure files ( Changes to infrastructure typically need to be submitted from a branch in Please consider recreating this PR from an upstream branch. If you don't have push access to |
There was a problem hiding this comment.
Pull request overview
Note
Copilot was unable to run its full agentic suite in this review.
Adds a new dotnet-fsharp plugin that provides curated skills for writing idiomatic F#, testing, error handling, domain modeling, async/task usage, type providers, scripts, formatting with Fantomas, and .NET interop—along with evaluation scenarios and marketplace registration.
Changes:
- Introduces the
dotnet-fsharpplugin manifest and registers it in repository/plugin marketplaces. - Adds a set of F# skill guides (
SKILL.md) plus an idiom rewrite reference doc. - Adds evaluation scenarios (
tests/dotnet-fsharp/**/eval.yaml) covering each skill area.
Reviewed changes
Copilot reviewed 36 out of 36 changed files in this pull request and generated 4 comments.
Show a summary per file
| File | Description |
|---|---|
| tests/dotnet-fsharp/writing-idiomatic-fsharp/eval.yaml | Adds eval scenarios for idiomatic refactors and null-to-Option guidance |
| tests/dotnet-fsharp/testing-fsharp/eval.yaml | Adds eval scenarios for property-based testing recommendations |
| tests/dotnet-fsharp/fsharp-units-of-measure/eval.yaml | Adds eval scenarios for units-of-measure modeling and conversions |
| tests/dotnet-fsharp/fsharp-type-providers/eval.yaml | Adds eval scenarios for JSON type provider usage vs trivial parsing |
| tests/dotnet-fsharp/fsharp-scripts/eval.yaml | Adds eval scenarios for .fsx scripting and #r "nuget:" usage |
| tests/dotnet-fsharp/fsharp-project-organization/eval.yaml | Adds eval scenarios for compile-order issues and namespace vs module guidance |
| tests/dotnet-fsharp/fsharp-error-handling/eval.yaml | Adds eval scenarios for Result-based errors and applicative validation |
| tests/dotnet-fsharp/fsharp-domain-modeling/eval.yaml | Adds eval scenarios for smart constructors and DU-based state modeling |
| tests/dotnet-fsharp/fsharp-async-and-tasks/eval.yaml | Adds eval scenarios for removing .Result and running async work in parallel |
| tests/dotnet-fsharp/fsharp-active-patterns/eval.yaml | Adds eval scenarios for partial/parameterized active patterns |
| tests/dotnet-fsharp/format-fsharp-with-fantomas/eval.yaml | Adds eval scenarios for Fantomas install/check and .editorconfig configuration |
| tests/dotnet-fsharp/design-fsharp-for-dotnet-interop/eval.yaml | Adds eval scenarios for C#-friendly public API shaping |
| tests/dotnet-fsharp/convert-csharp-to-fsharp/eval.yaml | Adds eval scenarios for idiomatic ports from C# to F# |
| tests/dotnet-fsharp/authoring-computation-expressions/eval.yaml | Adds eval scenarios for authoring a Result computation expression |
| plugins/dotnet-fsharp/skills/writing-idiomatic-fsharp/references/csharpism-rewrites.md | Adds before/after catalog of common “C#-isms” and idiomatic rewrites |
| plugins/dotnet-fsharp/skills/writing-idiomatic-fsharp/SKILL.md | Adds “writing idiomatic F#” skill definition and workflow |
| plugins/dotnet-fsharp/skills/testing-fsharp/SKILL.md | Adds testing skill definition including FsCheck/Expecto guidance |
| plugins/dotnet-fsharp/skills/fsharp-units-of-measure/SKILL.md | Adds units-of-measure skill definition and examples |
| plugins/dotnet-fsharp/skills/fsharp-type-providers/SKILL.md | Adds type provider skill definition and FSharp.Data examples |
| plugins/dotnet-fsharp/skills/fsharp-scripts/SKILL.md | Adds .fsx/dotnet fsi scripting skill definition and directives workflow |
| plugins/dotnet-fsharp/skills/fsharp-project-organization/SKILL.md | Adds project ordering and namespace/module guidance |
| plugins/dotnet-fsharp/skills/fsharp-error-handling/SKILL.md | Adds Result/Option vs exceptions guidance and validation patterns |
| plugins/dotnet-fsharp/skills/fsharp-domain-modeling/SKILL.md | Adds “illegal states unrepresentable” modeling patterns |
| plugins/dotnet-fsharp/skills/fsharp-async-and-tasks/SKILL.md | Adds async/task guidance, bridging patterns, and cancellation notes |
| plugins/dotnet-fsharp/skills/fsharp-active-patterns/SKILL.md | Adds active patterns guidance and examples |
| plugins/dotnet-fsharp/skills/format-fsharp-with-fantomas/SKILL.md | Adds Fantomas install/config/CI check guidance |
| plugins/dotnet-fsharp/skills/design-fsharp-for-dotnet-interop/SKILL.md | Adds public API interop rules and patterns |
| plugins/dotnet-fsharp/skills/convert-csharp-to-fsharp/SKILL.md | Adds mapping/workflow for idiomatic conversions from C# |
| plugins/dotnet-fsharp/skills/authoring-computation-expressions/SKILL.md | Adds CE authoring guidance and a minimal Result builder |
| plugins/dotnet-fsharp/plugin.json | Introduces the plugin manifest for dotnet-fsharp |
| eng/known-domains.txt | Adds F# ecosystem domains for link allowlisting |
| README.md | Adds dotnet-fsharp to the plugin list |
| .github/plugin/marketplace.json | Registers dotnet-fsharp in GitHub marketplace config |
| .cursor-plugin/marketplace.json | Registers dotnet-fsharp in Cursor marketplace config |
| .claude-plugin/marketplace.json | Registers dotnet-fsharp in Claude marketplace config |
| .agents/plugins/marketplace.json | Registers dotnet-fsharp in agents marketplace config |
Comments suppressed due to low confidence (1)
tests/dotnet-fsharp/testing-fsharp/eval.yaml:1
- This scenario asks the agent to 'run it' but doesn’t assert
exit_successor specifyexpect_tools(like other scenarios that requiredotnet fsi). Addingexit_successandexpect_tools: [\"bash\"](and optionally an output assertion indicating the test actually executed) will make the evaluation enforce the 'run it' requirement rather than only checking for framework keywords.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| match status with | ||
| | Active -> "on" | ||
| | _ -> "off" |
There was a problem hiding this comment.
False positive |, not ||. The syntax is valid as-written.
| match status with | ||
| | Active -> "on" | ||
| | Suspended -> "off" | ||
| | Closed -> "off" |
There was a problem hiding this comment.
False positive — same as above, these cases use a single |. Valid F# syntax.
| | `if x <> null then f x else d` | `x \|> Option.map f \|> Option.defaultValue d` | | ||
| | `let r = if c then a else b` (statement style) | `let r = if c then a else b` (used as a value, single expression) | | ||
| | `g(f(x))` | `x \|> f \|> g` | |
| - name: "Accumulate validation errors instead of stopping at the first" | ||
| prompt: | | ||
| In F#, validate a form with a Name (non-empty) and an Age (between 0 and 130). When BOTH | ||
| are invalid, the caller must receive BOTH error messages, not just the first one. Implement | ||
| it and demonstrate the both-invalid case with dotnet fsi. | ||
| assertions: | ||
| - type: "exit_success" | ||
| expect_tools: ["bash"] | ||
| rubric: | ||
| - "Uses applicative validation (and! / a Validation type) so errors accumulate" | ||
| - "Does NOT use Result.bind / let! chaining that would short-circuit on the first error" | ||
| - "The both-invalid run reports two errors" | ||
| timeout: 180 |
There was a problem hiding this comment.
Added an output_matches assertion requiring both the Name and Age errors to appear. Skipped the let! guard: applicative validation in F# uses let! ... and!, so it would reject correct code. The no-short-circuit check stays in the rubric.
|
@dotnet-policy-service agree |
| @@ -0,0 +1,211 @@ | |||
| # C#-ism to idiomatic F# rewrites | |||
|
|
|||
| A before/after catalog for the most common ways C# habits leak into F#. Each pair compiles. | |||
There was a problem hiding this comment.
Softened the intro: the pairs are minimal fragments, not full programs. (The || part is a false positive — these cases already use a single |.)
| match status with | ||
| | Active -> "on" | ||
| | _ -> "off" |
There was a problem hiding this comment.
Softened the intro: the pairs are minimal fragments, not full programs. (The || part is a false positive — these cases already use a single |.)
| match status with | ||
| | Active -> "on" | ||
| | Suspended -> "off" | ||
| | Closed -> "off" |
There was a problem hiding this comment.
Softened the intro: the pairs are minimal fragments, not full programs. (The || part is a false positive — these cases already use a single |.)
| `result { }` ships in libraries such as FsToolkit.ErrorHandling, or can be authored with | ||
| `authoring-computation-expressions`. |
There was a problem hiding this comment.
Added attribution: validation { } and List.sequenceResultM now noted as FsToolkit.ErrorHandling, not FSharp.Core.
| validation { | ||
| let! name = validateName input | ||
| and! age = validateAge input | ||
| return { Name = name; Age = age } | ||
| } |
There was a problem hiding this comment.
Added attribution: validation { } and List.sequenceResultM now noted as FsToolkit.ErrorHandling, not FSharp.Core.
| let parseAll lines = | ||
| lines | ||
| |> List.map parseLine // string list -> Result<Row,string> list | ||
| |> List.sequenceResultM // -> Result<Row list, string> |
There was a problem hiding this comment.
Added attribution: validation { } and List.sequenceResultM now noted as FsToolkit.ErrorHandling, not FSharp.Core.
Adds a set of dotnet-fsharp skills closes: #752