🎮 The Next Input — Issue #041

Your Team's Private AI Podcast

Aaron Bost
August 27, 2025

In partnership with

⚡ The Briefing — 60 sec

Anthropic launches a Claude AI agent that lives in Chrome. It was bound to happen!
Google Translate adds AI-powered language-learning features. Google vs. the Duolingo owl—my 日本語 might finally benefit!
Gemini’s AI image model gets a bananas upgrade. Hence the wacky photo for today’s issue 🍌

🛠️ The Playbook — Browser Agent Field-Ops (Claude-in-Chrome)

Mission Turn any open browser tab into an actionable workflow: extract → summarise → act (file a ticket, draft an email, update a record).
Difficulty Medium | Build time 60–90 min
ROI ICs and leads save ≈ 8–12 h/week on “read, copy, paste, action” glue work.

Step	Action
1	Trigger (Chrome Agent) – On hotkey, capture page DOM, selected text, and URL.
2	Normalise – Agent cleans boilerplate, resolves pagination, detects entities (company, contact, dates).
3	Summarise – Prompt: “Give 5 bullets: What happened, Why it matters, Owner, Deadline, Next step.”
4	Decide – Lightweight rules: if page is issue/PR → open Jira ticket; if it’s a lead → push to HubSpot; if it’s a spec → post to Notion.
5	Act via Webhooks (Make/Zapier) – Send structured JSON `{title, summary, entities, url}` to the right flow.
6	Confirm – Agent shows a one-tap preview (edit before send).
7	Fail-safe – If confidence < 0.8 or no entity match → save draft to “Needs Review” and Slack the owner.

Pro tip: Keep a tiny Routing Table (Google Sheet): domain_pattern → action. Updating rows adds new automations without touching prompts.

🗺️ Side Quest — Daily Briefing Podcast

This Week’s Side Quest Question
“We’re fully remote. Async stand-ups are walls of text no one reads. Can AI make a short, personalized ‘daily briefing’ podcast every morning?”

Remote teams everywhere know the pain: async stand-ups turn into walls of text, quickly skimmed (or skipped entirely). The result? Missed blockers, invisible shout-outs, and no shared focus. The right format needs to be short, human, and easy to consume while grabbing coffee—which is why a daily audio brief works.

The Core Concept: a 90–120s private podcast that surfaces context, highlights top priorities, and calls out blockers—delivered automatically to Slack or RSS.

Lean Stack (ship this week): Jira/Linear + Slack + Calendar → Make/Zapier aggregator → LLM generates script → TTS (PlayHT/ElevenLabs/Polly) → Slack MP3 upload or private RSS.

The “News Director” Prompt (mini rules): lead with top-3 updates (owner + deadline), then blockers, add 1–2 shout-outs, and finish with a crisp CTA. Keep it ≤120s, plain English, with SSML pauses for pacing.

Personalisation Options:

By Team – e.g., Sales Daily vs Eng Daily, filtered segments.
Segmented Master – One master script spliced into variants by role.
Per-Listener Tail – 30–45s appended for leaders/ICs: today’s meetings + one owned blocker.

V1 Build:

Trigger at 7:30am → fetch yesterday’s Jira/Slack updates + today’s calendar.
Deduplicate stories by title hash.
Send to LLM with the “News Director” prompt.
Convert script → MP3 (via TTS).
Post file to #daily-briefing in Slack.
Add 🔥/🧊 emoji for feedback.

Pro-Tips: add SSML <break> tags and brand stings for polish; normalise audio (-16 LUFS); expand acronyms on first mention; log skips to refine length.

🎯 The Arsenal — Tools & Prompts

Asset	What it does	Link
Make.com	Router for agent→SaaS actions	https://make.com
Raycast	Hotkey launcher for “send tab to agent”	https://www.raycast.com
ElevenLabs / PlayHT	TTS engines for natural audio	https://elevenlabs.io · https://play.ht
Prompt · One-Tap Action Card	Convert page → JSON summary/action

From this page, output JSON {title, summary(3 bullets), entities, recommended_action, owner, deadline, url}. Keep it factual and concise.

💡 Free Office Hours

Need a Chrome-agent router or daily-briefing bot?
Book a free 15-minute Office Hours slot—no sales pitch, just workflows solved.
→ https://calendly.com/aaron-cylentis/the-next-input-office-hours

Discover the measurable impacts of AI agents for customer support

How Did Papaya Slash Support Costs Without Adding Headcount?

When Papaya saw support tickets surge, they faced a tough choice: hire more agents or risk slower service. Instead, they found a third option—one that scaled their support without scaling their team.

The secret? An AI-powered support agent from Maven AGI that started resolving customer inquiries on day one.

With Maven AGI, Papaya now handles 90% of inquiries automatically - cutting costs in half while improving response times and customer satisfaction. No more rigid decision trees. No more endless manual upkeep. Just fast, accurate answers at scale.

The best part? Their human team is free to focus on the complex, high-value issues that matter most.

👉 Curious how they did it? Read the full case study to learn how Papaya transformed their customer support

🕹️ Game Over

Ship one tab→action flow today—tomorrow’s “copy-paste chores” disappear.
Got a win? Hit reply; you could headline Issue #042.

— Aaron
Automating the boring. Amplifying the brilliant.
Forwarded this? Subscribe → https://cylentisai.beehiiv.com/subscribe