# How we monitor brand mentions

A daily agent that searches news, social platforms, and forums for new mentions of our brand, classifies sentiment, and posts a digest of the notable or negative ones to Slack — each with a suggested response drafted for a human to review.

Canonical page: https://openchatly.com/use-cases/brand-monitor

Brand mentions show up wherever people happen to be talking — a news writeup, a
Reddit thread, a tweet, a forum post comparing us to a competitor. Most days
nothing needs a response. Some days a single negative post with real reach needs
one within the hour, and the team doesn't find out until it's already been seen
by everyone else.

We run a brand-monitor agent on Kortix that checks the web every day for new
mentions, classifies each one's sentiment, and posts a digest to our marketing
Slack channel. It never replies or posts anywhere itself — for anything notable,
it drafts a suggested response and leaves the sending to a person.

- **Team:** Kortix
- **Runs on:** Daily cron, reusable session
- **Connected systems:** The web · Slack
- **Mode:** Search-only · draft, never post

## The problem

Brand mentions aren't confined to one platform. They land in news coverage,
social posts, and forum threads on no predictable schedule, and most of them
don't matter — a passing mention, a neutral comparison, a repost. The ones that
do matter are indistinguishable from the noise until someone actually reads them,
and by the time a person gets around to checking, a negative post with real reach
has often already had its worst hour.

The usual workarounds don't fix this. A Google Alert fires on keyword matches
with no sense of sentiment or reach, so every alert looks the same regardless of
whether it's a five-word tweet or a widely shared negative review. Checking
manually catches yesterday's mentions today, if someone remembers to look. Either
way, nothing tracks what's already been seen, so the same viral post gets flagged
again tomorrow.

## What we built

A daily cron re-prompts a persistent agent session. Each run searches the web for
new mentions of the brand across news, social platforms, and forums, checks each
one against a ledger of mentions it has already reported, classifies the sentiment
of anything new, and posts a digest to Slack — a one-line rollup for routine
mentions, and a full callout with a suggested response draft for anything notable
or negative. It never posts, replies, or comments anywhere; the only output is the
Slack digest.

## How it works

### Copy the setup prompt into a new session

Open your Kortix workspace, start a new session, paste this prompt and send it.
The agent writes the agent file, the brand-mention-monitor skill and the daily
trigger, and opens a change request for you to review. Prefer clicks? Choose
**Use this template** instead.

**Setup prompt**

```
Set up a daily brand monitor agent in this workspace.

1. Create the agent agents/brand-monitor.md. Its job is to find new mentions of our brand on the public web and post a daily digest. It may use web search and fetch, and the Slack connector.

2. Create the skill skills/brand-mention-monitor/SKILL.md with these rules:
   - Search news, social platforms and forums for the brand terms I give you: brand name, aliases, product names and common misspellings.
   - Keep a ledger of every reported mention in memory/brand-monitor-log.md, and never report the same mention twice.
   - Classify each new mention as positive, neutral or negative. Mark it notable when it is negative, has high reach, comes from an influential source, or states a wrong fact about us.
   - For each notable mention, draft one suggested response and label it as a draft. Never reply, comment or post anywhere.
   - Post one digest to #marketing per run: a one-line rollup for routine mentions and a full callout for notable ones.

3. Add a cron trigger daily-brand-monitor to kortix.yaml. It runs the brand-monitor agent every day at 06:00 with the prompt "Load the brand-mention-monitor skill and run today's brand-monitor sweep."

4. Tell me which of these connectors are not connected yet.

5. Open a change request with all of it. Do not merge it.
```

### Connect your tools

The agent searches the public web with its built-in search and fetch, so the
sources need no connector. Open **Connectors** in your workspace and connect
Slack for the digest. Credentials are brokered server-side and never enter the
agent's machine.

- **Slack** (Required):
Posts the daily digest with suggested responses.
- **Reddit** (Optional):
Searches threads for new brand mentions, read-only.

### Run it once with this prompt

Merge the change request, then start a session with the **brand-monitor**
agent and send this prompt. The first run fills the ledger, so check what it
logged before you let the schedule take over.

**Prompt**

```
Run one brand-monitor sweep for our brand terms.

1. Read memory/brand-monitor-log.md for mentions already reported.
2. Search news, social platforms and forums for each brand term.
3. Skip every mention already in the ledger. Classify each new one by sentiment and notability.
4. Draft one suggested response for each notable mention. Do not post it anywhere.
5. Post one digest to #marketing, then add today's mentions to the ledger.
```

### Let the daily schedule take over

The change request added this trigger to `kortix.yaml`. From now on the agent
runs every day at 06:00 and picks up from its ledger in `memory/`. To change the time, edit the cron line, or open
**Triggers** in the workspace.

**kortix.yaml**

```yaml
triggers:
  - slug: daily-brand-monitor
    name: Daily brand monitor
    on:
      cron: "0 0 6 * * *"
    run:
      agent: brand-monitor
      prompt: Load the brand-mention-monitor skill and run today's brand-monitor sweep.
```

### Draft responses, never post them

The agent only searches and reads public pages, and it writes only to one
Slack channel. A suggested response appears in the digest as a labeled draft.
A person reads it, edits it and decides whether to send it.

### Read the digest

Each run ends with one digest in the channel. This is what a run looks like
in the session:

> **Daily brand monitor · 06:00**
>
> Prompt: Load the brand-mention-monitor skill and run today's brand-monitor sweep.
>
> Found **15** new mentions: **9** positive, **4** neutral and **2** negative.
> Flagged **2** as notable, a forum thread with a pricing complaint and a news
> article that misstates a feature. Each has a drafted response for your review.

> **The pattern**
> A daily **cron** re-prompts a reusable **session** that searches the web for
> brand mentions, dedupes against a **memory** ledger of what's already been
> reported, classifies sentiment, and posts a Slack digest. Anything notable
> gets a suggested response — drafted, never sent.

## Make it do more

Once it works, paste any of these into the same session.

**Track a launch**

```
For the next 14 days, add our new product name to the brand terms and post a separate launch digest each evening.
```

**Weekly sentiment trend**

```
Every Monday, post the count of positive, neutral and negative mentions for each of the last 8 weeks to #marketing.
```

**Route bug reports**

```
When a mention describes a product bug, also post its link and excerpt to #support.
```

## Guardrails

The agent watches public conversation about the brand, so its access is scoped
to searching and reading, with a single, narrow output:

- **Isolation.** Every run happens in its own isolated sandbox, and only the
  Slack digest it's explicitly allowed to send is written back out.
- **Scoped secrets.** Any credential the agent uses is encrypted in the secrets
  manager and injected into the sandbox at runtime, scoped to the agents you grant them to
  or the logs.
- **Draft, never post.** The agent's only external action is one Slack digest
  per run. It never replies, comments, or posts on the platform where a mention
  appeared — a suggested response is a draft for a human to review and send
  themselves.
- **Everything is code.** The brand terms, the watch scope, and the sentiment
  and notability rules are files in the repo, versioned and changed through a
  reviewed **change request** rather than a dashboard setting.

## The outcome

- **Every day:** New mentions found without anyone remembering to check
- **Dedupe by design:** Already-reported mentions never resurface
- **Draft only:** A suggested response, never a reply the agent sends itself

The team no longer finds out about a negative mention secondhand, and no one
re-reads yesterday's viral post because it was flagged twice. The channel gets
one digest a day — quiet on quiet days, specific and actionable when something
needs a response — and every response that goes out is one a person chose to
send.
