spb/llm-api-toolkits

1# openai-api-toolkit23An Agent Skill for writing, reviewing, debugging and migrating code that calls the OpenAI API.45Its model tables and request linter were built by sending 55 request shapes (Responses API and6Chat Completions) to each of 29 OpenAI text models and recording what the API accepted. The7linter's verdict matched the API on all 1,594 recorded outcomes.89Not affiliated with OpenAI. "OpenAI" and "GPT" are trademarks of OpenAI.1011## Contents1213| Path | Purpose |14|---|---|15| `SKILL.md` | When to use the skill, the working loop, the rules that cause most failures |16| `references/models.md` | Ids, context windows, output caps, prices (generated) |17| `references/request-rules.md` | What each model accepts on each API, with exact error texts (generated) |18| `references/requests.md` | Responses anatomy, streaming, structured outputs, files, embeddings, batches |19| `references/tools-and-agents.md` | Function calling, the loop, hosted tools, the Agents SDK |20| `references/state-caching-cost.md` | Conversation state, prompt caching, cost levers |21| `references/errors.md` | Status codes and observed rejections with fixes |22| `scripts/openai-api.mjs` | Offline CLI: `models`, `lint`, `fix`, `convert`, `migrate`, `cost`, `compare`, `scaffold`, `guide` |2324The scripts use only the Node.js standard library, open no network connection and need no key.2526## Install2728```bash29skills install @spboucher/openai-api-toolkit --client claude-code   # or codex, cursor, generic-agent-skills30```3132## Measured effect3334See `CHANGELOG.md` for a with/without-skill evaluation against the live API, with its caveats.3536## Limits3738A snapshot taken on the date in `SKILL.md`. Only text models visible to the probing key were39probed. Computer use, Realtime, fine-tuning, Assistants and Azure OpenAI were not probed. Cost40output is an estimate. When the API and this skill disagree, the API is right.41