Discussion topic relationships: six recorded examples
The downloads contain six original inputs, their expected labels and one recorded API observation per input revision. Disagreements stay visible.
Supply summaries of both conversations. Compare the same topic, a narrower subtopic and connected or unrelated discussions before a separate summary or merge decision.
The middle levels keep useful distinctions visible. Related discussions can inform one another without describing the same work; a subtopic is a concrete part or example of the broader conversation.
This page returns a label for one pair. It does not calculate a complete clustering solution, combine message history or write a merged summary. Those steps need application policy and additional evidence.
This original pair-comparison demo adapts the five relationship levels published in signal-summarizer, linked by the source post. It uses Choice labels instead of reproducing the upstream Score workflow and adds an explicit missing-summary outcome. No chat is merged.

Use the sample to understand the rule, then test your own input for discussion topic relationships.

Supply summaries of both conversations.

Open the decision rules and keep the intended distinction between Same, Subtopic, Related, Domain, Unrelated. Edit the choices only when testing your own rule.

Use a sample first, then try your own input. Compare the selected label with the other returned probabilities.

Review Unclear before using a result. The label is a decision about the supplied information, not an executed action.
| Original input | Observed label | Leading probability |
|---|---|---|
| Direct continuation | Same | 93.0% |
| Narrow example | Subtopic | 99.0% |
| Connected but distinct | Related | 94.0% |
| Only broad domain | Unrelated | 64.0% |
| No topic connection | Unrelated | 97.0% |
| Summary absent | Unclear | 100% |
These are six original synthetic inputs, tested once per request revision. The percentage is the leading option probability, not measured accuracy. Authored expectations and any disagreements remain in the downloadable evidence. 5 of 6 results matched the authored fixture expectations. For Only broad domain, the expectation was Domain; the observed choice was Unrelated.Verified
The downloads contain six original inputs, their expected labels and one recorded API observation per input revision. Disagreements stay visible.
The single-request SDK runner uses your private TYPESAFE_API_KEY. The offline consumer reads saved observations, checks request hashes and prints a pending output. It does not call the model or change any external system.
# Offline: inspect the downloaded saved observations. No API key is needed.
python preview_choice_workflow.py samples.json saved-api-results.json
# Optional live request: install the SDK and set TYPESAFE_API_KEY privately.
npm install @typesafe-ai/sdk@0.6.0
node run-choice-sample.mjs samples.json case-1
# The live response prints to the terminal; it does not update saved-api-results.json.The input contract and these limits define what the result means.
No. The demo only labels a pair. A grouping application must decide how to handle message history, permissions and other related threads.
Domain captures a very broad overlap such as software development. Related requires a more specific connection while preserving distinct discussion goals.
It describes the distribution over the supplied choices for this request. These six synthetic examples are functional tests with authored expectations, not a production accuracy benchmark.
No. The demo displays a label and distribution. The downloadable consumer produces a pending preview packet. Application code and a separate authorization policy would control any real action.
Edit the example above to see how a changed fact affects the decision, or open the Playground for another task.