generating-api-sdks
Tools to build idiomatic API client libraries in multiple languages from OpenAPI files.
Install
mkdir -p .claude/skills/generating-api-sdks && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/4703" && unzip -o skill.zip -d .claude/skills/generating-api-sdks && rm skill.zipInstalls to .claude/skills/generating-api-sdks
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Generate client SDKs in multiple languages from OpenAPI specifications.Key capabilities
- →Generate type-safe client SDKs from OpenAPI 3.0+ specifications.
- →Produce idiomatic client libraries in multiple languages.
- →Implement authentication handling for various schemes.
- →Add automatic retry logic with exponential backoff.
- →Build pagination helpers for cursor-based and offset pagination.
How it works
This skill parses an OpenAPI specification to generate typed models and client classes in various languages. It includes features like authentication, retries, and pagination, and generates documentation and test suites.
Inputs & outputs
When to use generating-api-sdks
- →Generating TypeScript client for frontend
- →Building Go SDK for backend services
- →Syncing API changes with client code
About this skill
Generating API SDKs
Overview
Generate type-safe client SDKs in multiple languages (TypeScript, Python, Go, Java) from OpenAPI 3.0+ specifications. Produce idiomatic client libraries with authentication handling, automatic retries, pagination helpers, and comprehensive type definitions that mirror the API contract.
Prerequisites
- OpenAPI 3.0+ specification file (YAML or JSON) with complete schema definitions
- Target language toolchain installed (Node.js/TypeScript, Python 3.10+, Go 1.21+, or Java 17+)
- OpenAPI Generator CLI or equivalent code generation tool
- Package registry credentials for publishing (npm, PyPI, Maven Central, pkg.go.dev)
- CI pipeline for automated SDK builds on spec changes
Instructions
- Read and validate the OpenAPI specification using Read, checking for complete schema definitions, proper
$refusage, and response type coverage across all endpoints. - Extract all operation IDs, request/response models, and authentication schemes from the specification to build the SDK method inventory.
- Generate typed model classes for every schema component, including nested objects, enums, and discriminated unions with proper nullability annotations.
- Create a client class with methods for each API operation, mapping operation IDs to idiomatic method names (e.g.,
listUsers,get_user,CreateUser). - Implement authentication handling with support for Bearer tokens, API keys (header and query), and OAuth2 client credentials flow, configurable at client instantiation.
- Add automatic retry logic with exponential backoff for 429 and 5xx responses, with configurable max retries and backoff multiplier.
- Build pagination helpers that abstract cursor-based and offset pagination into iterator/generator patterns native to each target language.
- Generate comprehensive JSDoc/docstring/Javadoc comments from OpenAPI
descriptionandsummaryfields for full IDE IntelliSense support. - Create a test suite that validates SDK methods against a mock server running the OpenAPI spec, covering authentication, error handling, and pagination.
See ${CLAUDE_SKILL_DIR}/references/implementation.md for the full implementation guide.
Output
${CLAUDE_SKILL_DIR}/sdk/typescript/src/client.ts- Main SDK client class with typed methods${CLAUDE_SKILL_DIR}/sdk/typescript/src/models/- TypeScript interfaces and type definitions${CLAUDE_SKILL_DIR}/sdk/python/client.py- Python SDK with dataclass models and async support${CLAUDE_SKILL_DIR}/sdk/go/client.go- Go SDK with struct types and context-based methods${CLAUDE_SKILL_DIR}/sdk/*/README.md- Per-language installation and usage documentation${CLAUDE_SKILL_DIR}/sdk/*/tests/- SDK test suites per language
Error Handling
| Error | Cause | Solution |
|---|---|---|
| Schema generation failure | OpenAPI spec contains $ref cycles or missing component definitions | Run spec linting with spectral first; resolve circular references with allOf composition |
| Type mismatch | API response does not match generated model types at runtime | Add runtime response validation option; log schema drift warnings for API provider notification |
| Auth configuration error | SDK instantiated without required credentials for protected endpoints | Throw descriptive error at client construction time listing required auth parameters |
| Pagination exhaustion | Iterator consumes all pages without termination condition | Enforce maximum page count safety limit; detect empty result sets as termination signal |
| Rate limit handling | SDK retry logic conflicts with application-level retry logic | Expose retryConfig option to disable built-in retries; emit retry events for observability |
Refer to ${CLAUDE_SKILL_DIR}/references/errors.md for comprehensive error patterns.
Examples
TypeScript SDK for payment API: Generate a fully typed PaymentsClient with methods like createCharge(amount, currency) returning Promise<Charge>, automatic Bearer token injection, and paginated listTransactions() returning an async iterator.
Python SDK with async support: Produce both sync (PaymentsClient) and async (AsyncPaymentsClient) clients using httpx, with Pydantic models for request/response validation and __aiter__ pagination support.
Multi-language CI pipeline: On OpenAPI spec change, automatically regenerate SDKs in all target languages, run tests against a mock server, bump semantic versions, and publish to respective package registries.
See ${CLAUDE_SKILL_DIR}/references/examples.md for additional examples.
Resources
- OpenAPI Generator: https://openapi-generator.tech/
- Speakeasy SDK generation: https://speakeasy.com/
- Microsoft Kiota for API client generation
- OpenAPI Specification 3.1: https://spec.openapis.org/oas/v3.1.0
Prerequisites
Limitations
- →OpenAPI spec must contain complete schema definitions.
- →OpenAPI spec must use proper `$ref` usage.
- →OpenAPI spec must have response type coverage across all endpoints.
How it compares
This skill automates the generation of feature-rich, idiomatic client SDKs from an OpenAPI spec, unlike manually writing client libraries for each API.
Compared to similar skills
generating-api-sdks side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| generating-api-sdks (this skill) | 1 | 25d | Review | Advanced |
| generate-subsystem-skills | 1 | 7mo | No flags | Advanced |
| copilot-sdk | 7 | 3mo | Review | Intermediate |
| replit-sdk-patterns | 1 | 25d | Review | Intermediate |
Try saying
Example prompts that trigger this skill in your AI assistant.
More by jeremylongshore
View all by jeremylongshore →You might also like
generate-subsystem-skills
llama-farm
Generate specialized skills for each subsystem in the monorepo. Creates shared language skills and subsystem-specific checklists for high-quality AI code generation.
copilot-sdk
github
Build agentic applications with GitHub Copilot SDK. Use when embedding AI agents in apps, creating custom tools, implementing streaming responses, managing sessions, connecting to MCP servers, or creating custom agents. Triggers on Copilot SDK, GitHub SDK, agentic app, embed Copilot, programmable agent, MCP server, custom agent.
replit-sdk-patterns
jeremylongshore
Apply production-ready Replit SDK patterns for TypeScript and Python. Use when implementing Replit integrations, refactoring SDK usage, or establishing team coding standards for Replit. Trigger with phrases like "replit SDK patterns", "replit best practices", "replit code patterns", "idiomatic replit".
mistral-hello-world
jeremylongshore
Create a minimal working Mistral AI chat completion example. Use when starting a new Mistral integration, testing your setup, or learning basic Mistral API patterns. Trigger with phrases like "mistral hello world", "mistral example", "mistral quick start", "simple mistral code", "mistral chat".
common-performance-engineering
HoangNguyen0403
Enforce universal standards for high-performance development. Use when profiling bottlenecks, reducing latency, fixing memory leaks, improving throughput, or optimizing algorithm complexity in any language.
mcp-builder
anthropics
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).