Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.tryflare.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
Quickstart

By the end of this guide, you’ll have connected your GCP account and run your first anomaly analysis.
No credit card required. Flare is free during the open beta.

​
Prerequisites

  • A Google Cloud Platform project with Cloud Audit Logs enabled
  • A Google account with roles/logging.viewer (read-only) on the project you want to analyze
1

Create your account

Go to tryflare.ai/sign-up and create your account. You’ll be redirected to the dashboard.
2

Connect GCP

Navigate to Connectors in the sidebar, then:
  1. Select Google Cloud Platform
  2. Click Continue through the provider info screen
  3. Click Connect with Google - this opens Google’s OAuth consent screen
  4. Sign in with your Google account and grant Flare read-only access to your Cloud Audit Logs
Once connected, you’ll see a green “GCP connected” confirmation with your Google email.
Flare requests only logging.read scope - it cannot modify your GCP resources.
3

Run your first analysis

Click Run analysis from the Connectors page, or go to Analyses > New Analysis. Then:
  1. Name your analysis - e.g., “Production audit - May 29”
  2. Select GCP Audit Logs as the source
  3. Enter your GCP Project ID - found in the GCP Console under the project selector
  4. Choose a time window - start with “Last 24h” for a quick first scan
  5. Click Fetch & Analyze
Flare will fetch your audit logs and run the analysis. This typically takes 15-60 seconds depending on log volume.
4

Review your findings

Once complete, you’ll see:
  • An AI summary describing the overall patterns detected
  • Ranked anomalies scored from 0-100, categorized as Critical, High, Medium, or Low
  • Source log evidence for each finding - the actual log entries that triggered the detection
  • A “First Seen” badge on anomalies involving values that have never appeared in your environment
Click any anomaly to expand it and see the baseline vs. query frequency comparison and raw source logs.
5

Ask follow-up questions

Use the Flare Intelligence chat panel on the right side of the results page. Flare has full context from the analysis and can answer questions like:
  • “Why is this anomalous?”
  • “What should I investigate first?”
  • “Is this a known attack pattern?”
  • “What are the remediation steps?”

​
What’s next

Set up scheduled runs

Automatically analyze your logs every 1h, 6h, or 24h

Upload log files

Analyze logs from any source by uploading JSON, CSV, or text files

Understand results

Learn how to read anomaly scores, categories, and evidence

GCP setup details

Permissions, troubleshooting, and reconnecting