---
name: "podcast-chapter-generator"
description: "Draft timestamped chapters from a supplied, timed transcript. Use when the user needs: Get chapter titles and start times with source evidence."
metadata: {"version": "2026-10-17.1", "category": "voice-audio", "publisher": "AIDiveForge"}
---

# Podcast Chapter Generator

Get chapter titles and start times with source evidence.

## Requirements

Provide a transcript with reliable timestamps and the target platform’s chapter rules.

## Workflow

1. Identify substantive topic transitions and retain their transcript timestamps.
2. Choose concise descriptive chapter titles rather than clickbait or unsupported summaries.
3. Keep starts in ascending order, remove duplicates and ensure every timestamp is within the episode duration.
4. Apply the supplied platform constraints; when none are supplied, return a neutral chapter list instead of claiming platform compliance.
5. Spot-check each boundary against audio when available and flag transcript-only uncertainty.

## Worked example

Read `examples/input.txt` and `examples/expected.txt` before adapting the workflow.

## Acceptance checks

Starts are ordered and below 12:00; titles reflect the supplied topics without inventing content.

## Limits

Transcript timing can drift. Platform acceptance and editorial quality require a final audio/platform check.

Worked example included. Model-specific performance and vendor compatibility have not been benchmarked.