Outreach reply classification: six recorded examples
The downloads contain six original inputs, their expected labels and one recorded API observation per input revision. Disagreements stay visible.
Compare a reply with the original outreach. Distinguish interest, a later follow-up, a redirect, an objection and automated responses, giving explicit stop-contact requests priority.
Different replies need different human follow-up decisions. An out-of-office message is not a rejection, and a ticket acknowledgment is not interest. Providing the original outreach also helps interpret short replies.
The Stop label is a safety exit in this example, not a complete contact-permission system. The page does not send mail, change a suppression list or grant permission to contact a referred colleague.
This original text demonstration adapts a decision in a repository linked by the cited X post. The examples and explicit uncertainty outcomes are ours. The post is a directory reference, not proof that its author built every linked project. No third-party benchmark, business outcome or external action is reproduced.

Use the sample to understand the rule, then test your own input for outreach reply classification.

Compare a reply with the original outreach.

Open the decision rules and keep the intended distinction between Interest, Later, Wrongperson, Objection, Away. 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 |
|---|---|---|
| Relevant interest | Interest | 100% |
| Later in the year | Later | 100% |
| Different owner | Wrongperson | 99.0% |
| Automatic absence | Away | 100% |
| Plain refusal | Objection | 95.0% |
| Stop despite praise | Stop | 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. 6 of 6 results matched the authored fixture expectations.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. It describes the reply. A separate process must check contact permissions, suppression rules and the appropriate timing before any action.
This page sorts an entire reply queue into several dispositions. The dedicated unsubscribe example examines opt-out meaning more closely; use it when that is the task you need to test.
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.