- Agents
- Cases
- Workflows
Open source security automation platform for teams and AI agents
Tracecat helps AI-native security teams build agents and automate work.
Wiz alert
wiz.finding.created
SOC analyst agent
Wiz MCP · Falcon MCP
Isolate host
falcon.isolate_host
Triage this Wiz finding: privilege escalation on a prod IAM role.
• Pull associated assets and CloudTrail context via Wiz MCP
• Correlate with EDR telemetry from Falcon MCP
I can route this through the SOC analyst agent — it will tune the finding into an actionable case, correlate EDR telemetry, and propose containment.
Yes — open a case and prepare containment.
Wiz (MCP connected)
CrowdStrike Falcon (MCP connected)Found Isolate host in Falcon. It will quarantine the endpoint and revoke active session tokens while you review.
Ready to run Isolate host on i-0a7f…. Approve to proceed.
Trusted by security builders replacing legacy SOAR
Scale your security workforce with agents you own.
Turn analysts into builders and builders into architects, so a team of 5 can do the work of 50.
Replace static workflows with custom agents →
GmailFalcon (EDR)
Okta
SlackTriage user-reported emails and contain credential harvest before more users click.
1. Parse headers and detonate links and attachments.
2. Pull Falcon telemetry for any user that clicked.
3. Quarantine endpoints and force a session revoke when credentials were entered.
Confirmed credential harvest. Two more users in Finance clicked the link in the last 90 minutes. Recommending session revoke and a tenant-wide block on the sender domain.
- Revoke active sessions for the 3 impacted users in Okta
- Block sender domain at the email gateway and Falcon DNS
Everything you need to make security agents work.
Automate work across your security stack with agents you control and customize.
Learn about Tracecat agents →Automate security work with prompts
Draft workflows, cases, tables, agent skills, and tool integrations from Claude, Cursor, or your AI workspace.
Skills built by your team
Reusable agent capabilities your engineers ship to GitHub. Analysts compose them into runbooks.
Hosted MCP servers
Connect agents to over 100 pre-built MCP servers without writing integration code.
Self-host anywhere
Deploy in your own VPC, on-prem, or on Tracecat's cloud. Your prompts and detections stay yours.
Human approvals on sensitive actions
Pause workflows for explicit approval before sensitive actions. Every approval is logged.
Audit every tool call
Open source audit logs of every prompt, tool call, and decision your agents make. In your Git, in your VPC.
Turn prompts into deterministic workflows.
500+ integrations across SIEM, EDR, MDM, IdP, and more.
Connect your coding assistant to Tracecat. Turn prompts into workflows.
Run loops, if-conditions, parallel subflows, and scripts (Python, Bash, Ansible).
Run agents in workflows with explicit tool approvals.
Trigger
New OAuth consent
Scatter
Split by user
Close alert
Allowlisted app
OAuth analyst agent
AI Agent
Revoke Google grant
Revoke Entra grantNotify ownerRun subflow
Account quarantine
Fully customizable case management with agents.
Reduce time-to-resolve with agents that collect evidence, summarize findings, and prepare the next action.
Explore Tracecat case management →Shai-Hulud npm worm — @ctrl/tinycolor exposure across 3 endpoints
Internal advisory flagged @ctrl/tinycolor@4.1.2 as compromised by the Shai-Hulud worm. A GitHub API search found references in 7 repos. CrowdStrike Falcon flagged the npm install and post-install script execution on 3 developer workstations.
Detection details
- GitHub advisory ingested at 14:08 UTC, Falcon detection at 14:42 UTC
- 34-minute detection delay
- Initial exposure via transitive dependency in package-lock.json
Indicators of compromise
- Malicious package: @ctrl/tinycolor@4.1.2
- Post-install SHA: a1f2c3…d4e5
- C2 callback: webhook.site/<redacted>
- Affected endpoints: WS-042, WS-118, WS-203
Recommended actions: pin affected repos to clean versions, rotate npm + GitHub PAT tokens, isolate affected endpoints in Falcon, and audit GitHub Actions runners for token reuse.
Summary drafted from the case description.
Suggested next steps
- Pin clean versions across 7 affected repos
- Rotate npm and GitHub PAT tokens org-wide
- Isolate WS-042, WS-118, WS-203 via Falcon
I can draft a containment runbook and pre-fill actions from the IOC list — token rotation, endpoint isolation, and repo owner notifications.
Drive Tracecat with your favorite coding assistant.
Define playbooks in natural language. Give everyone the power to automate security work through Tracecat MCP.
Read the Tracecat MCP docs →
Works with your favorite coding assistant.
Explore examples of security agents and workflows.
See how teams automate their security work with the tools they already use. Each example is a starting point an agent builds around your stack. They cover every control in NIST CSF 2.0.
OktaOkta
Microsoft Entra ID
AWSTracecat is mission ready.
Learn about the Enterprise Edition →Talk to a Tracecat expert
Loved by security teams building with AI
Security Engineer @ Depop
Tracecat copilot has changed my life. I describe an agentic workflow and it builds it for me. I never had time to build and experiment around my other responsibilities. Now I do.
Senior Security Engineer @ Neo Financial
A genuine thank you to the team. I built an end-to-end IoC enrichment pipeline with Claude and Tracecat MCP and created more value for our SOC in a day than I probably would have in weeks on my own. You're making my one-man SOC assignment possible.
Principal Threat Researcher @ Saronic
Tracecat is a cheat code for corporate security teams that want to build and own their own agentic future.