Nim Api Design logo

Nim Api Design

Community
planetis-m
nim-api-design

Design clear public Nim APIs for libraries and modules, including exported types, constructors, parameter ownership, lookup functions, error contracts, and container-style interfaces. Use when creating or reviewing a Nim library API, designing an exported module surface, or deciding how callers should construct, pass, access, and validate data.

Overview

Publisherplanetis-m
Repositoryskills
Skill namenim-api-design
Stars
66
Forks
7
Bundled files
6
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 6 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by planetis-m on GitHub. Read the source before you install it.

Installation

Install the Nim Api Design AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/planetis-m/skills/tree/master/nim-api-design
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/planetis-m/skills.git /tmp/skills
mkdir -p .claude/skills
cp -r /tmp/skills/nim-api-design .claude/skills/nim-api-design
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Nim Api Design in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Nim Api Design on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Nim Api Design is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Nim API Design

Rules

Public shape

  • Prefer plain object types for public data models.
  • Use ref object only when identity, aliasing, shared mutation, graph structure, or handle lifetime is part of the contract.
  • Export one primary public representation per concept. Add paired value/ref APIs only when both forms are part of the contract.
  • Prefer procs, overloads, generics, and iterators. Do not default to methods or runtime dispatch for ordinary APIs.
  • Use named object types for public semantic data.
  • Use tuples only for local glue or iterator yields such as (key, val).
  • Reuse stdlib names when the behavior matches: len, find, contains/hasKey, [], []=, items/mitems, pairs/mpairs, add, del, clear, incl/excl, and push/pop.
  • When indexed removal must preserve sequence order, use delete(idx), not del(idx).
  • Use find for an index or position result. Use contains or hasKey for boolean membership.
  • For collection-like types, expose items and pairs. Add mitems and mpairs only when callers may safely mutate yielded values.
  • For comparable types, define ==, < and <=. Do not define !=, >, or >=; Nim derives them.

Contracts

  • Use range types for constrained public parameters; store their values in the underlying type (int for Natural or Positive).
  • Pass arguments directly to range parameters; conversion is implicit, so do not write Positive(x).
  • Model closed choice sets as enums; add reserved members for future assignments in bounded protocol code spaces; use strings for open-ended or externally extensible token namespaces.
  • When an enum's string spelling is dictated by an external format or protocol, set explicit strings (dirNorth = "north"); $ and parseEnum[T] always round-trip.
  • Use static[T] in public APIs only when callers must supply a compile-time constant.
  • Bare typedesc parameters may name different types; share T across typedesc[T] parameters when they must match.
  • Use distinct when two values share a base type but must not mix.
  • If a public type can be used as a table or set key, define hash consistent with ==.
  • Use func for pure query operations when purity is part of the public contract.
  • Use {.raises: [].} when a proc must not raise. Leave raising procs unannotated.

Constructors and conversions

  • Use initX() for value types that copy by value.
  • Use newX() for ref types and for value types with reference semantics.
  • Use one toX() name for common conversions. Overload on input type. Do not define converters.
  • Use default parameter values for optional constructor arguments, keeping the no-argument call the main path.

Sequence parameters

  • When a parameter must accept both arrays and seqs, use openArray[T] for read-only traversal or var openArray[T] for fixed-length element mutation.
  • Resizing or replacement requires var seq[T]: an openArray is fixed-length.
  • When you control the callers, name the concrete type (array[N, T] or seq[T]) instead of openArray[T].

Parameter ownership

  • Use T when the caller's variable stays unchanged, var T when the proc changes it, sink T when the proc takes ownership, and lent T only for borrowed returns.
  • Passing x: T never copies what x points to. Do not use ptr T or addr x to avoid a copy.
  • Pass sink arguments normally — do not wrap them in move() or ensureMove(). Last-use auto-sink moves locals, their fields, tuple fields and indices, and direct-indexed seq/array elements with no copy.
  • Use move(x) only where auto-sink does not apply: var parameters and their fields, loop-indexed seq/array elements (arr[i]), and values reused later in the same scope.
  • Use ensureMove(x) to make an accidental copy a compile-time error.

Template parameters

  • Treat template parameters as syntax substituted into the template body. Design them around the expressions, names, types, or statement blocks the caller must provide—not like proc parameters.

Lookup surface

  • Separate required lookup from optional lookup.
  • A required lookup raises one specific catchable exception. It does not return a silent default.
  • Use contains or hasKey for membership checks and getOrDefault for explicit fallback values.
  • If several accessors fail the same way, route the failure through one private {.noinline, noreturn.} helper.

Borrowed and mutable access

  • Use lent T for read accessors that return storage owned by the receiver.
  • Return var T only for deliberate mutable views. If changes require validation or related updates, expose mutation procs instead.
  • In lent and var accessors, return directly from storage. Do not route through a temp local.

Public boundary

  • Export only the stable surface. Keep helpers private.
  • Keep accessor-backed fields private, so external dot access routes to the accessors; an exported field would bypass them.
  • Use descriptive public names.
  • In user code, gate version-specific API with when (NimMajor, NimMinor) >= (x, y).
  • Keep template lookup escape hatches such as withValue secondary to the main lookup surface.

Workflow

  1. Choose the representation. Start with a plain object. Switch to ref object only if the contract needs identity or aliasing.
  2. Name the public types. Use named objects for semantic results and distinct types for domain identities.
  3. Name the constructor surface. Use initX, newX, and toX in the stdlib style.
  4. Design the lookup surface. Provide one strict path for required data and one explicit safe path for optional data.
  5. Choose parameter modes and borrowed access. Use T when the caller's variable stays unchanged, var T when the proc changes it, sink T when the proc takes ownership, and lent T to return a borrow. Use mutation procs when changes need validation or related updates.
  6. Verify the contract. Compile public examples. Exercise each mutation, ownership, and failure path.

Common Mistakes

MistakeWhy it is wrong
Starting with ref object for plain dataIt adds aliasing and shared mutation where the API does not need them
Defaulting to methods or runtime dispatchIt hides behavior behind runtime polymorphism when a proc surface is simpler and clearer
Weakening a constrained public parameter to int and re-checking manuallyIt throws away a stronger boundary contract
Returning a silent default for required dataIt hides missing-data bugs
Returning var T for controlled stateCallers can bypass validation and related updates
Returning a lent or var result through a temp localThe temp dies at return; the compiler rejects the dangling borrow
Using lent T for an input parameterlent T is a borrowed return type; the compiler rejects it in parameter position
Assuming a sink call always moves the caller's variableNim copies the argument when it cannot prove last use
Wrapping a routine sink argument in ensureMoveSink already performs last-use analysis; use ensureMove only when the code must fail instead of copy
Defining !=, >, or >= for a typeThese override Nim's derived comparison templates and can make comparisons inconsistent
Assuming variant of branches scope their own fieldsReusing a name is a compile error

References

  • references/representation_and_construction.md — Choosing value or reference semantics with initX and newX construction.
  • references/object_variants.md — Variant branches sharing attribute names through per-branch payload objects.
  • references/lookup_and_mutation.md — Collection access through required, optional, borrowed, and controlled-mutation paths.
  • references/parameter_and_result_shapes.md — Domain IDs, related options, constrained parameters, and named results in one public API.
  • references/parameter_ownership.md — Choosing openArray, var, sink, and lent parameters.
  • references/template_usage.md — Templates for scoped cleanup, caller-named access, and lazy evaluation that a proc cannot express.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Nim Api Design AI skill do?

Design clear public Nim APIs for libraries and modules, including exported types, constructors, parameter ownership, lookup functions, error contracts, and container-style interfaces. Use when creating or reviewing a Nim library API, designing an exported module surface, or deciding how callers should construct, pass, access, and validate data.

Why use Nim Api Design on TypingMind?

Because you install it once and use it with any model. Nim Api Design is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Nim Api Design in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/planetis-m/skills/tree/master/nim-api-design. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Nim Api Design?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Nim Api Design?

As many as you like. As long as a model supports skills, you can use Nim Api Design with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Nim Api Design AI skill free?

It is published on GitHub by planetis-m. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇