Personal data in logs: six recorded examples
The downloads contain six original inputs, their expected labels and one recorded API observation per input revision. Disagreements stay visible.
Show the logged fields and how the logger or redactor handles them. Check whether personal values reach the output or whether the implementation needs closer review.
The log format alone does not reveal every value it will contain. A nested object or serializer can carry personal fields into output even when the visible message is generic.
Redacted or nonpersonal needs positive implementation evidence. A function name such as safeLog is not enough, and an unknown identifier should remain unknown rather than being declared safe or personal by spelling.
The source semcheck rule asks whether selected log calls write personal data. This original example adds explicit uncertainty handling and uses synthetic fields only; it does not audit a repository or certify a logging system.
This is an independently authored text demonstration inspired by the linked X discussion. Its rules and synthetic inputs are ours. It does not reproduce the author’s application, verify reported outcomes, or perform an external action.

Use the sample to understand the rule, then test your own input for personal data in logs.

Show the logged fields and how the logger or redactor handles them.

Open the decision rules and keep the intended distinction between Exposed, Redacted or nonpersonal, Review. 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 Review before using a result. The label is a decision about the supplied information, not an executed action.
| Original input | Observed label | Leading probability |
|---|---|---|
| Email written directly | Exposed | 97.0% |
| Verified redaction | Redacted or nonpersonal | 98.0% |
| Service event only | Redacted or nonpersonal | 99.0% |
| Nested personal address | Exposed | 99.0% |
| Unknown helper | Review | 94.0% |
| Ambiguous identifier | Review | 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. Its meaning and context matter. This example asks for the supplied field definition instead of inferring it from the name id.
No. It reads the behavior you provide. A real review should verify that description against the actual implementation and tests.
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.