Resources and prompts
Resources live in src/Presentation/Resources, prompts in src/Presentation/Prompts, and the completion handler in src/Presentation/Completions. In a solution created with --include-auth, all of them need a token. Every resource and prompt has an icon, a movie, an author, a review or the server, from the brand repository.
Resources
Section titled “Resources”Every resource returns application/json. The movie, author and review resources are templates, so they’re listed under resources/templates/list; the top rated and version resources are plain resources under resources/list.
| URI | Name | Returns |
|---|---|---|
movies://top-rated |
top_rated_movies |
A page of the ten best rated movies, highest average stars first |
movies://{movieId} |
movie |
A movie with every review |
authors://{authorId} |
author |
An author with every review |
reviews://{reviewId} |
review |
A review |
server://version |
server_version |
The version stamped into the build |
{ "fileVersion": "1.1.0.0", "informationalVersion": "1.1.0.0+c7e22dee4ffd508b82caca702ebb713d0a2ba8bc" }An unknown id fails with the JSON-RPC -32002 resource not found error, and an id that isn’t a GUID fails with -32602 invalid params.
Prompts
Section titled “Prompts”| Prompt | Arguments | What it asks the model to do |
|---|---|---|
summarise_movie_reception |
movieTitle |
Summarise a movie’s average stars, how its ratings spread, and who liked it most and least |
summarise_movie_reviews |
movieId |
Summarise what reviewers said about a movie: the themes they agree on, where they disagree, and whether their words match their stars |
compare_movies |
firstMovieTitle, secondMovieTitle |
Compare how reviewers rated two movies and say which they preferred |
summarise_author_reviews |
authorName |
Summarise an author’s reviewing habits: how many reviews, how generous they are, and the movies they rated highest and lowest |
review_movie |
movieTitle, authorName |
Find the movie and the author, ask for a star rating and what they thought, and save the review with reviews_write |
summarise_movie_reception and compare_movies look the movie up by title, and summarise_author_reviews looks the author up by name. When they find it, by an exact match or as the only match, the returned messages embed it as a movies:// or authors:// resource, reviews and all, so the model has the data without another call. When they don’t, the messages tell the model which tools to use to find it. summarise_movie_reviews takes the movie’s id, so it always embeds the movie, or fails with the JSON-RPC -32002 error when there’s no movie with that id.
Completions
Section titled “Completions”The movieTitle, firstMovieTitle and secondMovieTitle arguments complete from movie titles, and authorName completes from author names, up to 50 at a time with the total count.
{ "completion": { "values": ["Deep Space Gardeners", "Postcards from Outer Space", "Space Station Bake Sale", "The Last Space Lighthouse"], "total": 4, "hasMore": false } }Ids complete too: the movieId argument of summarise_movie_reviews, and the ids in the movies://{movieId}, authors://{authorId} and reviews://{reviewId} templates. Type the start of an id, or part of the movie’s title or the author’s name, and you get the matching ids. A review is found by its movie’s title.
{ "completion": { "values": ["cb188ef6-14e7-f85f-2f78-924c2ab26b28", "dfefa450-5454-6a4d-a4aa-95a9b047c17b", "ec850910-9692-cf5d-8ce4-938b304e9228", "ec68dc8c-2c08-5139-b68f-6c4be5061bdb"], "total": 4, "hasMore": false } }A value over 100 characters, or an argument the handler doesn’t know, completes to nothing.