| id | course-mcp-tools-l04 | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| type | course | |||||||||||||||
| title | The input_required round-trip | |||||||||||||||
| summary | A tool can ask the client a question mid-call — return the requests, finish on the retry carrying answers. | |||||||||||||||
| lang | en-US | |||||||||||||||
| content_version | 1 | |||||||||||||||
| status | reviewed | |||||||||||||||
| reviewed_on | 2026-09-12 | |||||||||||||||
| hints |
|
A tool can ask the client a question mid-call — return the requests, finish on the retry carrying answers.
The 2026-07-28 replacement for server-initiated elicitation: a handler raises InputRequired(requests); the server returns resultType "input_required" with the questions; the client retries the same call with params.inputResponses; the handler finishes. Implement it, then verify like an engineer: both verifier directions, the full suite, and a diff review where every change is forced by a test.
- Hand-trace one round-trip with pen and paper before running anything
- Write the three-line evidence note: verified / not verified / known limits
The lesson's commands run, and you can answer these in your own words (the agent asks; you answer — that is the gate):
- Which command did this lesson run, and what did it decide?
- What failed at the start, and why — in your own words?
- Before trusting a similar change next time, what would you check first?
The command transcript and your answers.