# Hearl

> Hearl collects what people say about your product on X, Reddit and Hacker News, filters out the noise with AI, and gives each post an owner, a deadline and a clear outcome.

Hearl gathers mentions from X, Reddit and Hacker News into one queue, filters out the noise, and tracks who replied, who's waiting and what still needs an answer.

## Your team saw it. Nobody owned it.

A bug report sits in a subreddit for three days. A buying question on X gets two likes and no answer. Someone shares a screenshot in Slack, and everyone assumes someone else replied. Listening tools tell you what people said. They don't make sure anyone answers.

## From a public post to a closed loop

### Collect

Hearl watches your official X accounts, the posts that mention them, your subreddits and site-wide Reddit search, and keeps every reply thread up to date.

### Filter

Hearl's AI reads each post against your product description and tells you whether it's really about you, whether it needs a reply and how urgent it is. Posts it isn't sure about go to a review list instead of disappearing.

### Own

Each conversation gets one owner and as many collaborators as it needs. Support can pull in engineering without losing the original post or the context.

### Close

Hearl keeps checking the thread. When your team replies, the status changes on its own, and when the user follows up, the conversation comes back to the inbox.

## What the inbox does for you

### Noise filtered before you look

Posts about a game engine function that shares your name, a stock ticker, a joke reply. Each judgment shows its probability, so you can see why something was kept or set aside.

### Reply status from the platform itself

Hearl reads the thread to see whether your team answered and whether the user wrote back. No one has to remember to tick a box.

### One owner, many collaborators

Claim a conversation, invite a colleague, add notes with @mentions, set a due date and record what was done.

### Built for keyboards

Move through the queue with j and k, claim with a, finish with e. Triage a morning's worth of mentions in a few minutes.

## One post often needs more than one team

### Support

Catch questions and bug reports where users actually post them, then hand off to engineering with the full thread attached.

### Product

Keep feature requests with their original wording and the replies that agree with them, so they reach planning as evidence.

### Engineering

See the user's setup, the support notes and the deadline in one place, without asking for the link again.

### Marketing

Spot criticism early, agree on a response with product, and keep a record of what was said.

## Your AI agent works the same inbox

Connect Claude, ChatGPT, Cursor or a script through MCP, the CLI or the API. It can triage posts, draft replies and write reports; every change shows in the timeline, and Hearl still never posts for you.

- Triage: "Triage my Hearl inbox and assign today's posts."
- Draft replies: "Draft replies to the unanswered Reddit posts about the sync bug."
- Follow-ups: "Which users followed up since yesterday, and who's on them?"
- Weekly report: "Write this week's community feedback report from Hearl."

See what agents can do: https://hearl.org/agents

## Questions

### What is Hearl?

Hearl is a shared inbox for public conversations about your product. It collects posts and replies from X, Reddit and Hacker News, uses AI to filter out posts that aren't about your product, and lets your team assign, discuss and resolve each one while it tracks the reply status on the platform.

### How is it different from a social listening tool?

Listening tools focus on volume, sentiment and trends. Hearl focuses on what happens after a post is found: who owns it, who is helping, when it's due, whether the team replied and whether the user is satisfied.

### Which platforms are supported?

X, Reddit and Hacker News. Each product can track official accounts, mentions, subreddits and search terms. Hearl collects them in the cloud at your plan's frequency, so there is nothing to install.

### Does Hearl reply on my behalf?

No. Hearl links you to the original post so your team replies with its own account, then it checks the thread to confirm the reply went out.

### How does the AI filtering work?

Hearl's AI judges each post against your product description and gives each judgment a probability. Posts above 70% are kept, posts below 30% are filtered, and anything in between goes to a review list for a person to decide.

### Can I try it before paying?

Yes. The free plan covers one product and two members, and every new workspace starts with 14 days of the Team plan.

---
Source: https://hearl.org/