A local, open-source Mac app for listening to your Claude Code and Codex agents as they work. Each agent gets its own station: tune in to one, or listen to a Global Mix across all active agents. Hear progress, blockers, decisions, errors, and attention requests in real time.
We built Agent FM because we constantly struggle with context switching between multiple agents. We usually end up with 6–10 coding agents running in parallel, and keep losing track of which one is blocked, which one is waiting on approval, and which one has quietly gone off the rails.
The goal is not to replace the terminal or IDE. We still use those. Agent FM is the ambient layer that helps us stay in the loop without reading every line of every terminal.
Agent FM observes local Claude Code and Codex sessions, distills noisy activity into higher-signal updates, and surfaces the useful parts:
what the agent is doing
when it seems blocked
what changed
when tests fail
when it needs attention
when it makes a visible assumption or decision
Agent FM is a free and open-source Mac app. Narration uses a bring-your-own-key model: configure Gemini or OpenAI, and usage is billed by that provider. Agent FM runs locally and does not proxy requests through a hosted service.
We built a lot of the later parts of Agent FM while listening to Agent FM itself. It was genuinely fun, and that was the moment the workflow clicked for us.
Please try it out and let us know how it maps to your workflow. We’re especially curious where the narration feels useful, where it feels too noisy, and which agent states you wish it understood better.
No comment highlights available yet. Please check back later!
About Agent FM for Claude Code & Codex on Product Hunt
“Tune in and stay in the loop with your agents 🎧”
Agent FM for Claude Code & Codex launched on Product Hunt on May 14th, 2026 and earned 58 upvotes and 1 comments, placing #75 on the daily leaderboard. A local, open-source Mac app for listening to your Claude Code and Codex agents as they work. Each agent gets its own station: tune in to one, or listen to a Global Mix across all active agents. Hear progress, blockers, decisions, errors, and attention requests in real time.
Agent FM for Claude Code & Codex was featured in Open Source (68.4k followers), Developer Tools (512.4k followers), Artificial Intelligence (468.5k followers) and GitHub (41.2k followers) on Product Hunt. Together, these topics include over 194.2k products, making this a competitive space to launch in.
Who hunted Agent FM for Claude Code & Codex?
Agent FM for Claude Code & Codex was hunted by Aniket Deshpande. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.
Want to see how Agent FM for Claude Code & Codex stacked up against nearby launches in real time? Check out the live launch dashboard for upvote speed charts, proximity comparisons, and more analytics.
Aniket Deshpande
Hey Product Hunt, Aniket here.
We built Agent FM because we constantly struggle with context switching between multiple agents. We usually end up with 6–10 coding agents running in parallel, and keep losing track of which one is blocked, which one is waiting on approval, and which one has quietly gone off the rails.
The goal is not to replace the terminal or IDE. We still use those. Agent FM is the ambient layer that helps us stay in the loop without reading every line of every terminal.
Agent FM observes local Claude Code and Codex sessions, distills noisy activity into higher-signal updates, and surfaces the useful parts:
what the agent is doing
when it seems blocked
what changed
when tests fail
when it needs attention
when it makes a visible assumption or decision
Agent FM is a free and open-source Mac app. Narration uses a bring-your-own-key model: configure Gemini or OpenAI, and usage is billed by that provider. Agent FM runs locally and does not proxy requests through a hosted service.
We built a lot of the later parts of Agent FM while listening to Agent FM itself. It was genuinely fun, and that was the moment the workflow clicked for us.
You can download Agent FM for macOS at https://agentfm.ai. The code is open source at https://github.com/agentfm-ai/ag....
Please try it out and let us know how it maps to your workflow. We’re especially curious where the narration feels useful, where it feels too noisy, and which agent states you wish it understood better.