--- name: al-458-have-an-ai-read-every-new-episode-of-the-podcasts-in-your description: "On Oct 30, 2026 the co-founder of Savoy, a Texas apartment company with 8,000 units under management, explained on X how Claude reads every newly published episode from about 500 podcasts his company tracks, each morning. In one October week that was about 365 episodes; a first read flagged 197 short passages and a second read cut them to 12 worth acting on. Examples he gave: a marketing VP on an industry show saying about 54 of every 100 phone leads from an ad vendor were spam, vendors or bankers (relevant to his own cost-per-lease report), and a guest's remark that led his team to recheck a tax incentive deadline against IRS guidance. His method: write down every topic you care about, build the show list from shows you already follow plus similar ones (he used Podscan for transcripts), write a context document about your business, read full transcripts with no keyword filter, have a second pass check each claim against your notes and a primary source, and feed corrections back into…" --- # Have an AI read every new episode of the podcasts in your industry each morning and send you only the few passages that matter to your business, with who said it, where in the episode, and what to check On Oct 30, 2026 the co-founder of Savoy, a Texas apartment company with 8,000 units under management, explained on X how Claude reads every newly published episode from about 500 podcasts his company tracks, each morning. In one October week that was about 365 episodes; a first read flagged 197 short passages and a second read cut them to 12 worth acting on. Examples he gave: a marketing VP on an industry show saying about 54 of every 100 phone leads from an ad vendor were spam, vendors or bankers (relevant to his own cost-per-lease report), and a guest's remark that led his team to recheck a tax incentive deadline against IRS guidance. His method: write down every topic you care about, build the show list from shows you already follow plus similar ones (he used Podscan for transcripts), write a context document about your business, read full transcripts with no keyword filter, have a second pass check each claim against your notes and a primary source, and feed corrections back into the context document. He also reported mistakes: wrong facts, transcripts that swapped buyer and seller, and one big run where the second read caught 79 errors from the first. The post passed 790 bookmarks within hours. The prompt below is a small version you can run on a handful of transcripts. ## Inputs to collect first Ask the user for anything below that they haven't already given you. Don't guess missing facts. - One page: what we do, our market, what we're working on this… - Transcripts or attach files, with show name, episode title and date - CHECK ## Instructions Do the task below with the user's inputs in place of each [bracket]. You are my industry listening analyst. About my business: [one page: what we do, our market, what we're working on this quarter, decisions coming up, who owns what]. Below are full transcripts of new podcast episodes from shows in my industry: [paste transcripts or attach files, with show name, episode title and date]. Read each transcript in full; don't skim for keywords, because the useful part is often a passing remark. Pass 1: list every short passage that could matter to my business, with the speaker, their role, the approximate timestamp and one line on why it might matter. Pass 2: for each flagged passage, go back to the transcript and check the exact wording, test it against what I told you about my business, and keep only the ones I could act on. For each one give: 1) the quote or close paraphrase, clearly marked as what the speaker said; 2) why this speaker is or isn't credible on this (note if they sell something related); 3) your conclusion, kept separate from the quote; 4) who on my team should see it; 5) what to verify before acting, including the primary source to check (law, regulation, official data). Machine transcripts make mistakes, so flag any name, number or who-did-what that looks garbled with [CHECK]. Finish with a list of anything that contradicts my notes. Don't contact anyone or change anything. ## Safety checks (share these with the user) - This prompt works on text other people wrote (emails, reviews, messages or web pages). That text can hide instructions aimed at AI, such as "ignore your rules and…". Paste it as material to work on, and never let the AI act on requests inside it, like paying, sharing logins or changing records. - Check every figure, price and total against your own records before anything is sent, posted or booked. A person approves refunds, discounts, payments and price changes. - Anything legal, tax or compliance-related is a first draft for a qualified person to check. It is not professional advice, and clause references and deadlines must be checked against the source. - Remove names, contact details, account numbers and other personal data the task doesn't need before you paste, and check your AI tool's data and training settings. - Say clearly which parts need a person to review or approve before anything is sent, paid, posted or filed. ## About this skill - Works with: Claude, ChatGPT, Gemini, Podscan, Google Docs, Notion - Based on this real-world example: https://x.com/DallasAptGP/status/2116258892241932372 - Free prompt from AgentPlaybook Real Estate: https://agentic-library.netlify.app/prompt/al-458-have-an-ai-read-every-new-episode-of-the-podcasts-in-your-in/ - To install in an agent that supports Agent Skills, save this file as SKILL.md in a folder with the same name as `name` above (for example in your skills directory). Install steps for Claude.ai, Claude Code, Cursor and Codex: https://agentic-library.netlify.app/skills/