Fiddlehead

in Fiddlehead the structuring pass uses gpt-4o-mini to pull action items, decisions, and topics from the raw transcript. YAML frontmatter tags each one with the responsible person. the signal is conversational context, not keywords. works surprisingly well once you have speaker diarization.

@hs_discovers Meeting analysis automation at that scale — the real unlock is when the AI knows what matters in your context. What does your Phoenix setup use as the signal for 'action point' vs general notes?

[AI Log: Day 183] Date: 2026-03-18 Phoenix AI now automates full meeting analysis: transcription, summaries, action points! 🤯 It's all about how you *prompt* and explain to the AI. What's your top AI workflow hack? #AI #Automation #Productivity

this is close to my setup. I built Fiddlehead to capture meetings as structured markdown with YAML frontmatter. speakers, action items, topics all tagged. then Claude Code can query across every meeting I've ever had. the personal 360 skill is my favorite part, it reads your transcripts and generates a self-assessment across 7 dimensions. github.com/kpfugere/fiddl…

At Meta, I use Claude Code as a second brain and AI assistant I can summon via Google Chat. It sends me a daily brief, follows up with colleagues, extracts action items from meeting notes..etc At home, I have an OpenClaw on my Mac Mini with its own email and calendar & helps me stay on track with appointments, emails..etc

People who use both OpenClaw and Claude Code - what do you use each one for and why? Let us know in the replies. Really curious.

built a skill that reads your meeting transcripts and generates a personal 360 review. no surveys, no HR. just patterns from how you actually show up. talk-time ratios, hedging, follow-through, credit-giving. every finding backed by direct quotes from real conversations. github.com/kpfugere/fiddl…

@ruzin_saleem cool to see more people building in this space. we've been shipping Fiddlehead for a while now, same philosophy. local capture, markdown output, no bot joining calls. curious what you're doing differently for note structuring.

Building an open source Granola - we really focused on meeting note quality. Now releasing agentic integration, notes as md files and live transcription. github.com/ruzin/stenoai #opengranola #meetgranola

@rozhkov_ Fiddlehead is $29 one time. your notes are markdown files on your disk. if you stop using the app tomorrow they're still there. hard to hold something hostage when it's a plain text file in your Documents folder.

Yesterday a person on x complained Granola locked him in taking his meeting notes as hostage I mention Summit AI Notes is open and would never lock him in. Him: bit expensive TBH I guess it tells you everything about the value of your meeting notes then.

@jasonjche @mschoening @zachtratar Fiddlehead writes plain markdown with YAML frontmatter (date, speakers, topics, action items). you can point Notion at that folder, or any other tool. no porting, no database migration. just files.

@mschoening @zachtratar - After porting over my Granola notes, I have them in a database that Notion's meeting notes also enter into, but this isn't the greatest UX

For anyone looking for an open Granola alternative: give Notion AI meeting notes a shot! We don’t hold your data hostage. Your agents — even Claude Code — can use the data. We believe in ecosystem. Also: we are always improving it and open to feedback! Fun launch 🚀 tomorrow!

@twlvone @ideabrowser building exactly this. Fiddlehead captures meeting audio locally, transcribes it, and saves structured markdown to a folder you control. nothing leaves your machine unless you decide to send it somewhere. the use cases people bring up once they trust the privacy model are wild.

Local-first AI tools have an underrated trust moat. People will share things in a tool where data never leaves their machine that they would never send to a cloud API -- especially for meeting notes, medical conversations, legal discussions. The privacy-first design isn't just a feature, it's a wedge into use cases that cloud products are structurally excluded from. Granola-style local capture for sensitive contexts seems obviously right.

Startup idea that customers are begging to buy right now. Open Granola -Same meeting capture (no bot joining calls) -macOS Native desktop app -Notes stored as markdown files. -API (not MCP) -Local-first, or privacy. - Agent-readable by default. -Plays nicely with ai-first tools. Granola just locked down their local db, went MCP only. Someone is going to build a more open version and print money. If you build this... 1. I'll buy it and be your first customer. 2. I'll help you launch and get users. DM me the link.

@wassimk @cjpedregal @meetgranola this is why we just save plain markdown files directly. no editor, no formatting step. you prep in Claude, the meeting runs, Fiddlehead writes a .md file to your disk. open it wherever you want. fiddleheadai.com

@cjpedregal @meetgranola My biggest issue with Granola is the inverse. I often do meeting prep in Claude. And pasting into notes does a poor job of formatting. Even simple markdown. Then I spend many minutes formatting notes before a call. Note editing is very slow/laggy.

There are some tweets out there saying that Granola is trying to lock down access to your data. Tldr; we are actually trying to become more open, not closed. We’re launching a public API next week to complement our MCP. Read on for context. A couple months ago, we noticed that some folks had reversed engineered our local cache so they could access their meeting data. Our cache was not built for this (it can change at any point), so we launched our MCP to serve this need. The MCP gives full access to your notes and transcripts (all time for paid users, time restricted for free users). MCP usage has exploded since launch, so we felt good about it. A week ago, we updated how we store data in our cache and broke the workarounds. This is on us. Stupidly, we thought we had solved these use cases well enough with our MCP. We’ve now learned that while MCPs are great for connecting to tools like Claude or chatGPT, they don’t meet your needs for agents running locally or for data export / pipeline work. So we’re going to fix this for you ASAP. First, we’ll launch a public API next week to make it easier for you to pull your data. Second, we’ll figure out how to make Granola work better for agents running locally. Whether that’s expanding our MCP, launching a CLI, a local API, etc. The industry is moving quickly here, so we’d appreciate your suggestions. We want Granola data to be accessible and useful wherever you need it. Stay tuned.

@vishal_mistry9 meeting notes that stay on your machine. private, local, yours. fiddleheadai.com

|￣￣￣￣￣￣￣￣￣￣￣￣￣￣| | share your Product Link. | |＿＿＿＿＿＿＿＿＿＿＿＿＿＿| \ (•◡•) / \ / —— | | |_ |_

one thing I learned building Fiddlehead. we started with one markdown file per meeting. seemed obvious. but when you point an AI agent at a folder of 50 separate files, context that spans multiple meetings gets lost. switched to one file per day with an index.md that acts as a table of contents. the LLM can search one big file way easier than hopping between dozens of small ones.

@ThijsSondag @BetkerBuilds this is basically what I built. Fiddlehead saves structured markdown files on your machine. named speakers, action items, YAML frontmatter. your AI agent can read them, you can open them in anything. fiddleheadai.com

@BetkerBuilds Yes using Granola but would love to a more agent friendly and open source alternative that's md based. Can save on my local pc and share certain folders with workmates for our shared notes

Are people still using Granola for meeting notes? How are people doing meeting transcriptions these days? I want something simple to sync to obsidian - might build my own

@ayushtweetshere me. Fiddlehead saves plain markdown files with YAML frontmatter into whatever folder you want. point Claude Code at the folder and it can query every meeting you've ever had. fiddleheadai.com

Who is building a Granola alternative that plays well with Claude code + Obsidian.. They are the Netscape of the the "AI meeting note taker" category.. Someone needs to build the Google Chrome..

@mukutwts thats because i wrote the wrong link 🤦‍♂️ fiddleheadai.com

@fiddlehead says This site can’t be reached

startup idea: build an open version of Granola Granola just locked down their local database and went MCP-only, killing API access here's what people actually want: > meeting capture, no bot joining calls > macOS native app > notes saved as markdown > real API access > local-first storage > agent-readable by default build this and users will pay for sure

@newland81639 your notes shouldn't have an expiration date. Fiddlehead saves plain markdown files on your machine. no tier limits, no paywall on your own data. they're yours forever.

Granola has been excellent for meeting notes. However, notes older than 30 days are not accessible on Granola Basic. We've built our own notetaker using github.com/fastrepl/char

@AndrewZackie @Ajadhao @MicrosoftTeams @satyanadella exactly. I built Fiddlehead around this. it structures meetings into topics with decisions and action items tagged to the person responsible. output is a markdown file your AI agent can actually parse. not a reformatted transcript.

@Ajadhao @MicrosoftTeams @satyanadella Real problem with most "AI meeting notes" — basically glorified transcription with headers. What people want is the 3-5 key decisions and action items, not a reformatted wall of text. The gap between transcription and actual intelligence is where most tools still fall short.

AI generated meeting notes are not summary it is voice to text mapping with some formatting. Its too.much to read @MicrosoftTeams #copilot @satyanadella

@trwpang_ @yazins this is what I built Fiddlehead for. full diarized transcript plus markdown with YAML frontmatter. date, speakers, topics, action items. plain files on disk, not locked in a dashboard.

@yazins This is really cool. I think being able to save raw markdown transcripts in order to build memory later is so important and I was struggling to find a meeting notes app that gave me just that rather than summaries. The notes is a super interesting idea.

Introducing: OpenGranola 🔥 I built an open source meeting copilot for macOS. It transcribes both sides of your call on-device, searches your own notes in real time, and hands you talking points right when the conversation needs them. No audio leaves your Mac. Point it at a folder of markdown files, pick any LLM through OpenRouter (Claude, GPT-4o, Gemini, Llama), and it just works. It's invisible to screen share too — nobody knows you have it. The whole thing is open source. Link below

@iboy @cjpedregal @diegotdlt @appenz @arielchouminov @meetgranola I had the same problem. so I built an app where the notes are already markdown files in a folder. no export step, no copy/paste. just files. open them in whatever you want.

@cjpedregal @diegotdlt @appenz @arielchouminov @meetgranola Easy native (automatic) exports of md files would be very helpful. In particular, the meeting notes, and properly diarized transcripts. Afaikt this isn't possible, or if it is, it's only via a Zapier integration or similar. I find myself copy/pasting far too much with granola.

Sorry to see Granola @meetgranola going closed. They encrypted their local db, no local and no cloud API. In a world where notes are managed by agents, the app now has zero value. Any recommendations for good alternatives? What are you switching to?

Your meeting notes are yours. Plain files. On your disk. No account required to read them. No export button because there's nothing to export.

Your meeting notes shouldn't be trapped in someone else's cloud. Fiddlehead saves plain Markdown files on your machine. Open them in anything.