The whole API in one file
Everything about the API in files you can hand to a tool or a language model. No key needed to read them.
The whole documentation as one Markdown file: every endpoint, field, error, plan and guide.
llms.txtA short guide for language models, linking everything else.
openapi.jsonOpenAPI 3.1: every endpoint, filter and field, for generating clients.
Postman collectionEvery call ready to run. Import it and set TT_KEY.
Give it to a language model
Paste the address of manual.md into Claude, ChatGPT or Cursor, or save the file into your project, and ask for the code you need. It is plain Markdown with every link written out in full, so nothing is lost when it is copied.
curl -O https://tabletennisapi.com/documentation/manual.md
Use with Claude and agents
The agent skill is one file that teaches an AI agent how to use the API: getting a key, the calls that answer the common questions, the limits, the errors and a typical workflow. It is built from the same source as these pages. Download skill.md, then save it as ~/.claude/skills/tabletennisapi/SKILL.md, or upload it as a skill in Claude's settings. Agents find it by themselves at /.well-known/agent-skills/index.json.
mkdir -p ~/.claude/skills/tabletennisapi curl -o ~/.claude/skills/tabletennisapi/SKILL.md https://tabletennisapi.com/skill.md
Generate a client
Any OpenAPI 3.1 generator can build a typed client from openapi.json. The examples in it, in the manual and in the Postman collection use real ids from a recent final, so every call works as written.
npx @openapitools/openapi-generator-cli generate \ -i https://tabletennisapi.com/openapi.json -g python -o tabletennis-client
Always current
All four files are built from the same source as these pages whenever it changes, and the coverage figures in the manual are counted from the data each day. Each file carries the date it was last updated.
openapi.json and the Postman collection. The manual also lists the planned ones, marked as not built yet.