Rolli Agent is an autonomous AI social intelligence agent that monitors social narratives across 8 platforms and 120,000+ news sources. It detects coordinated inauthentic behavior, scores narrative authenticity, and delivers structured intelligence briefs to your team or API endpoint — automatically, without manual intervention.

Rolli IQ AgentsNEW

Claude-powered · 94.2% detection accuracy

The AI social intelligence agent. Set it. Get briefed. Autonomously.

Rolli Agent monitors 8 platforms and 120K+ news sources, surfaces coordinated threats in real time, and delivers structured intelligence briefs to your team or API endpoint — no dashboards, no query language, no analyst required.

Deploy Your Agent →Request a Demo

Included in Rolli IQ Pro at $99/mo · Standalone agent API from $199/mo · No CC required

Natural language · no query syntax 8 platforms · real-time Autonomous 24/7 monitoring

Live investigation · right now

IQ vs Agent — which is right for you?

Agent investigation is included in Rolli IQ Pro at $99/mo. Standalone API access starts at $199/mo.

Rolli IQDashboard

You investigate. The dashboard provides the data. Best for analysts who want to drive their own research.

No-code dashboard

Manual, interactive investigation

For comms and security teams

$99/mo· free trial included

Explore Rolli IQ →

You are here

Rolli AgentAutonomous

Agentic investigation in natural language. For power users and developers who need answers, not dashboards.

Natural language — no query syntax

Autonomous 24/7 investigation

For power users and developers

$199/mo· free trial included

Keep reading →

Rolli APIDeveloper

Build authenticity scoring into your own product pipeline. Structured JSON, REST endpoint, no dashboard required.

REST API with structured JSON output

Integrate into your own tools

Best for dev teams and platform builders

Explore the API →

The difference

Search vs. Agent: a different class of answer

Every other tool gives you data. Rolli Agent plans an investigation, draws conclusions, and tells you what to do.

Traditional search

1.Run a keyword search

2.Sift through raw results

3.Open analyst tools separately

4.Build your own hypothesis

5.Write the brief yourself

✗→ 2–4 hours later

Rolli IQ Agents

1.Ask your question in plain English

2.Agent plans the investigation

3.Tool calls execute automatically

4.Signals scored for authenticity

5.Structured brief delivered to you

→ Under 3 minutes

Why it's different

Not a search. An investigation.

Every other tool gives you data and asks you to draw conclusions. Rolli IQ Agents draws the conclusion, shows the evidence, and tells you what to do next.

Speaks plain English, not SQL

Ask the question you'd ask your best analyst. No query language, no configuration, no dashboards to navigate.

Plans its own investigation

Sequences tool calls, pulls data from multiple endpoints, and synthesizes findings — the way a senior researcher would.

Tells you what to do

Every response ends with a recommendation. Escalate to Legal. Hold comms. Monitor 48 hours. The answer you need to act.

Evidence-backed, not opinionated

Every claim is grounded in signal data — authenticity scores, cluster maps, account behavioral fingerprints.

01 — Live now

47 accounts flagged. Coordinated. Right now.

Rolli IQ spotted it before any analyst did — 47 accounts posting identical language across X, Reddit, and Facebook in synchronised bursts. This is what always-on looks like.

Keywords

Users

Insights

Reports

Agents

Monitor

Dr. Robert Williams

History

Briefing Planner

Pinned

No pinned runs

Today

Tesla recall — consumer reaction analysis

2h · completed

American Airlines OTP — Q1 social sentiment

5h · completed

Southwest disruption — passenger reaction

1d · completed

Connected to RolliIQ

flagged

Suspicious accounts detected

Coordinated activity across 3 platforms — right now

Active Investigations

2 runs · 1 running

Running5m

Airline industry sentiment — 30-day trend

Searching signals…

Completed2h

Travel industry crisis tracker — coordinated actors

Ask a follow-up or start a new run…

Activity

Signals

Usage

Evidence

No active run

Try Rolli IQ Agents free

No credit card required

How it works

How the Rolli Agent works

No setup. No training data. No configuration. The autonomous AI social intelligence agent runs its own investigation — from question to structured brief in under 3 minutes.

Ask in plain language

Type any question — crisis, competitive, research, or monitoring. No query syntax, no filters to configure.

Agent plans the investigation

Rolli IQ Agents decides which tool calls to make, in what order, across which platforms. You see every step as it runs.

Agent scores and synthesizes

Signals are scored for authenticity, coordination is detected, clusters are mapped, and a structured brief is drafted.

You get an answer + recommendation

Plain-language response with evidence, confidence score, and a specific recommendation. Then ask your follow-up.

Watch it work

The agent in action

Collecting signals, running analysis, detecting coordination patterns — all in under 3 minutes.

Insights

Reports

Agents

Monitor

Completed

Query

"What is the social reaction to our CEO's climate comments in the past 24 hours?"

Completed · 91% confidence

Agent Activity893 signals

callCollecting posts from X, Reddit, BlueskyPast 24h · CEO climate keywords

resultSignal retrieval complete893 posts · 4 clusters

callRunning sentiment analysisScoring tone and amplification source

resultSentiment analysis complete71% negative · activist accounts

callScanning for coordinated patternsAccount age, timing, network fingerprints

resultCoordinated cluster confirmed31 accounts · synthetic posting pattern

61% of signals

27% of signals

Bluesky

12% of signals

Results

Sentiment71% negative

893

posts scanned

flagged accounts

Coordinated cluster detected — 31 accounts, synchronized posting

Try Rolli IQ Agents freeNo credit card required

Sample agent output

One question. One comprehensive brief.

Rolli IQ Agent · Intelligence BriefLIVE

PROMPT
“What's driving the narrative around [Brand X]'s CEO on social this week? Is it organic?”
AGENT BRIEF · GENERATED IN 28 SECONDS
Narrative velocity↑ 340% vs 7-day avg — spike began Tue 14:22 UTC
Authenticity score31/100 — HIGH COORDINATION SIGNAL
Origin pattern73 accounts created within 6 days · 89% zero-prior-activity
Platform spreadTwitter/X (primary) → Reddit (amplification) → Telegram (coordination)
RecommendationDo not respond publicly. Prepare holding statement. Alert legal.

New — Automated delivery

Scheduled IQ Briefs

Set the agent to run on your schedule — daily, weekly, monthly, or any custom cadence. Every morning your team wakes up to a structured intelligence brief in their inbox or Slack, with no one having to ask.

Any cadence you choose

Daily morning briefs, weekly narrative recaps, monthly trend summaries — or a fully custom schedule. You control the frequency.

Email or Slack delivery

Briefs land directly in your team’s inbox or a dedicated Slack channel — formatted, structured, and ready to forward to leadership.

Same depth as on-demand

Every scheduled brief is a full investigation: authenticity scoring, coordination detection, and a recommendation — not just a keyword summary.

Insights

Reports

Agents

Monitor

3 briefs scheduled

1 · Configure

2 · Investigate

3 · Deliver

New Brief Setup

Topic / search query

CEO climate statement coverage — past 24h

Cadence

Daily

Weekly

Real-time

Delivery time

08:00 AM

Channels

EmailSlack

Agent Activity

Collecting signals — X, Reddit, LinkedIn

08:00:12

Scoring sentiment & authenticity

08:00:31

Scanning for coordination patterns

08:00:47

Generating structured brief

08:01:02

—

Brief Delivered

Delivered to inbox · 08:01 AMSent

Tue, Mar 24 — Daily Brief

CEO Climate Statement Coverage · 893 signals

Sentiment71% negative

31 accounts — coordinated amplification cluster

Next deliveryWed, Mar 25 · 08:00 AM

Active Briefs

CEO climate coverage

Daily · 08:00 AM

Competitor narrative watch

Daily · 07:30 AM

Industry sentiment

Weekly · Mon 09:00

Next Up

Runs at08:00 AM

Delivers toEmail + Slack

Est. signals800–1,200

Set it and forget it

Choose topic, cadence, and delivery channels. Takes 30 seconds.

Schedule your first brief free

No credit card required

The output

One question. A complete intelligence brief.

Every Rolli Agent response includes signal findings, authenticity scoring, coordination detection, and a concrete recommendation — not a data dump.

Insights

Reports

Agents

Monitor

Completed · just now

CompletedHigh confidence · 94%

Competitor AI Launch — Social Narrative Analysis

Today, Mar 23 · 2:41 PM · 2,156 signals · X, Reddit, LinkedIn

Key Findings

4 findings

Overall sentiment: 61% negative — benchmark integrity concerns dominate

Top claim: Independent tests show inflated accuracy scores

Suspicious activity: 31 accounts posting near-identical methodology critiques

Amplification: 3 known short-sellers in coordinated cluster

Coordinated activity detected

31 accounts amplifying identical benchmark critique language in synchronised bursts — 3 are linked to known short-seller networks. Activity began 4 hours after the launch announcement.

"Their published benchmarks use cherry-picked datasets. Independent replication shows 23% accuracy gap on real-world tasks."

Rolli IQ Analysis

The 61% negative sentiment is real — but it's driven by two separate dynamics. Organic scepticism from ML practitioners (about 1,600 posts) reflects legitimate questions about benchmark methodology, and that's the story you'd expect at any high-profile AI launch.

The other 31 accounts are a different problem. Synchronised burst timing, near-identical phrasing, and links to financial short-sellers suggest this is an adversarial campaign, not grassroots concern. The coordination started within hours of the launch — likely pre-planned.

Recommendation: respond to the organic scepticism with methodology transparency (open test datasets, third-party audit). Do not engage with the coordinated cluster publicly — document it for legal and flag it to platform trust & safety.

Based on 2,156 verified signals · 31 suspicious accounts flagged

Overall Sentiment

2,156 posts analysed61% negative

61%

Negative

27%

Neutral

12%

Positive

Platform Distribution

X / Twitter1251 · 58%

Reddit625 · 29%

LinkedIn280 · 13%

Narrative Clusters

HIGHBenchmark integrity claims634

31 accounts · coordinated

HIGHMissing enterprise features498

MEDIUMPricing vs. value concerns387

LOWPositive demo reactions266

MEDIUMIntegration gaps197

1/6 · Report header

Instant context — no setup

Topic, confidence score, signal volume, and platforms — delivered in seconds.

Follow up:

Ask a follow-up question…

Try Rolli IQ Agents freeNo credit card required to start

Customer results

What teams say after their first Rolli Agent run

“I wish we would have had this YEARS ago. It could spare people countless hours of sifting.”
AR
Aimee Rinehart
Senior Product Manager, AI Strategy · The Associated Press
< 3 min
to first answer

“Rolli is top of the heap when it comes to narrative intelligence for communications teams. The authenticity scoring alone puts it in a category of its own, and the AI summaries are a favorite. Makes the report much easier to digest.”
HP
Head of Product
Major PR Distribution Platform
0 config
no query language

“Never has a tool such as this been so valuable with the influx of disinformation spreading on social channels. Rolli is a game-changer. Absolutely essential.”
Mike Reilley
Data and Digital Journalism · University of Illinois Chicago
Game- changer
disinformation defense

Access

Included in Rolli IQ Pro.

No separate product to buy. Rolli IQ Agents is part of the Rolli IQ dashboard at $99/mo — the same plan your team already uses for narrative monitoring.

Rolli IQ ProMost popular

$99/mo

Rolli IQ Agents — full access
5 free IQ credits/mo
Scheduled IQ Briefs — email or Slack
8-platform signal monitoring
Authenticity scoring — all signals
Evidence brief export
Follow-up conversation threads

Start Free Trial

Rolli IQ EnterpriseFor large teams

Custom

Unlimited investigation credits
Team-wide agent access
Scheduled IQ Briefs — custom cadence
Custom investigation templates
SSO + enterprise key management
DPA + compliance docs
Dedicated customer success

Talk to Sales

Product Proof

What the API actually returns

Real endpoint. Real schema. Authenticity scoring, coordination flags, and velocity data — structured for your product pipeline.

rolli-api · narratives endpoint · live

Request
POST /api/v1/narratives HTTP/1.1
Authorization: Bearer rk_live_•••••••••••••

Response200 OK— 47ms
{"topic": "Q2 earnings narrative",
"velocity_score": 8.4,
"authenticity": 0.34,
"coordinated_flag": true,
"origin_accounts": 14,
"platform_coverage": ["x", "reddit", "threads", "bluesky"],
"brief": "67 accounts showing coordinated amplification...",
}
Webhook delivered · Cluster map ready · Brief exportableGet API Keys →

Eval keys issued in 48 hours. Production SLA available. From $0.65/call.

FAQ

Common questions about Rolli Agent

Everything you need to know before you deploy your first agent.

What is Rolli Agent?

How does Rolli Agent differ from Rolli IQ?

What platforms does Rolli Agent monitor?

How is Rolli Agent different from a regular Rolli IQ search?

How does it know when something is coordinated vs organic?

Is there an API for Rolli Agent?

Deploy Your Agent →

Free trial · No credit card required

30-second intelligence

From question to intelligence brief in under 30 seconds.

No query language. No dashboards. Just ask.

Deploy Your Agent →

Your AI social intelligence agent is ready to deploy.

Start autonomous monitoring across 8 platforms and 120K+ news sources. First structured brief in under 3 minutes. No credit card, no configuration required.

Joined this week by 47 communications, security, and research teams

Deploy Your Agent →Request a Demo

Join 400+ organizations protecting their reputation. Average setup: 8 minutes.

Monitoring live in under 2 minutes · No credit card · Cancel anytime · SOC 2–aligned