---
name: tbt-project-fact-risk-prediction
description: "Ask for labels tied to exact wording and avoid treating forecasts as events for Separate a project's facts, risks and predictions with AI. Use permitted source material; this kit does not authorise external actions."
---

# Separate a project's facts, risks and predictions with AI

Ask for labels tied to exact wording and avoid treating forecasts as events.

Use the supplied source as authority. Check the account supports the actual file or image operation before attempting it.

## Task instruction

Map the claims in these supplied project notes. For each, retain exact
wording and source ID, date if supplied, and classify it as reported
event, current/pending status, risk, forecast/proposal or unresolved.
Keep source-reported facts separate from independent verification.
Preserve conditions and uncertainties. Do not convert expected dates
into commitments or infer a completed action from a plan or permission.
Then draft a clear update using the same bounded claims.

## Source boundaries

- Reported staging check
- Approval may delay release
- 15 October proposed conditionally

## Review and correction

A reported fact is not independently verified here. Risk and prediction should not become observed outcomes.

The update says [completed/committed claim]. Source [ID] establishes
[reported fact or conditional proposal]. Restore the correct status,
date condition and evidence boundary. Check related timeline labels
and summary headlines too. Keep the useful concise explanation and
do not invent an approval or test result to close the gap.

Use references/practice-source.json for the labelled example. The practice HTML is a local teaching tool; it does not call an AI service. Apply this method to the user's supplied facts rather than copying fictional example values. Do not infer approval or permission to publish.