Build a personal podcast digest that downloads authorized episodes, transcribes them locally, avoids duplicate work, and turns each transcript into a consistent research brief with traceable quotes, themes, and company mentions.
Download the skill for this workflow. Version 1.0.1.
with Claire Vo
Tomasz shows a local podcast processor that downloads episodes, converts and transcribes the audio, cleans the text, tracks processed files in DuckDB, and produces a daily digest with summaries, quotes, investment ideas, and company mentions.
A dated digest that covers each new episode once, links every extracted item to its episode and timestamp, and stores enough provenance to revisit the original context.
Follow the sequence once, then adapt the prompts, checks, and handoffs to your own setup.
6 steps
Read each permitted feed or input list, identify new episodes with a stable ID, download the audio, convert it with ffmpeg, and transcribe it with Parakeet or Whisper. Save the original metadata and timestamped transcript together.
Clean filler and obvious transcription noise without shortening ideas, rewriting quotes, or removing technical detail. Keep timestamps and preserve the raw transcript beside the cleaned version.
Clean this timestamped podcast transcript for readability. Remove filler and obvious transcription artifacts while preserving meaning, sequence, technical language, named entities, and every timestamp. Do not summarize, improve quotations, or add missing words. Mark uncertain passages as [unclear]. Transcript: [text].
Run the pipeline on a schedule and use DuckDB or another small database to record episode identity, source, status, model versions, output paths, and errors. A rerun should skip successful unchanged work and retry failed stages safely.
Generate a structured research brief from the cleaned transcript. Require citations to the episode timestamp for every quote, company, claim, and actionable observation.
Create a research brief from this transcript for [role or research goal]. Include a concise summary, key topics, notable claims, exact quotes, possible implications, and named companies. For every item, include the episode title and timestamp. Distinguish transcript evidence from your inference. Transcript: [text].
Make the schema explicit so every daily digest is scannable and comparable. Include host, guest, source link, publication date, summary, themes, quotes, implications, company mentions, and follow up questions, omitting empty sections rather than inventing content.
You are an expert analyst summarizing a podcast for a venture capitalist. Analyze the following podcast transcript and generate a structured summary. Transcript: [paste cleaned transcript here] Structure your output with the following sections, exactly as specified: - Host and Guest: - Comprehensive Summary: - Key Topics and Themes: - Actionable Quotes: - Investment Theses: - Noteworthy Observations: - Company Mentions:
For each company mention, return the exact transcript sentence, timestamp, and any identifying context. Confirm the organization before enriching a research record, and keep outreach or investment decisions outside the automated pipeline.
Extract organization mentions from this transcript. Return the name exactly as spoken, timestamp, surrounding sentence, likely canonical organization only when supported, and confidence. Do not infer contact details or create outreach. Transcript: [text].
Turn an idea into a PRD, user stories, and a plan.
How to recover when the loop fails and where human judgment helps.
Connect data from your hobbies (like golf scores or chess games) to an AI agent to get automated analysis, visual dashboards, and actionable coaching insights. Turn raw data into a story that helps you improve.
Turn your unstructured Instagram 'Saved' collection into a fun, personalized shopping catalog. This workflow uses an AI agent to organize your saves and present them in a nostalgic, engaging format like the SkyMall magazine.
Resurrect your old digital content, like a forgotten blog from Blogger or Tumblr, by using an AI agent to find, consolidate, and transform it into a daily personal podcast. Reconnect with your past self in a new, listenable format.
Join 250,000+ product managers who use ChatPRD to write better docs, align teams faster, and build products users love.
