AI-powered investigation. Human-led decisions.
Multi-Layer AI™
Your analysts.
Better prepared.
Help your team understand an alert, gather the surrounding evidence and decide what to do next. Keep every finding open to human review.
Multi-Layer AI™
From a warning
to a clearer decision.
Give your team a head start: gather context, connect the evidence and prepare the investigation for review. Available checks depend on your connected services and AI configuration.
A security alert starts the investigation.
Example investigation with synthetic data. This animation does not access your systems or take action.
- Less manual gatheringBring relevant context into the investigation.
- Clearer handoversKeep observations and findings together.
- Decisions you can explainReview the evidence before approving a response.
An alert. A place to start.
A suspicious attempt to access credentials arrives from your SIEM. SOCPilots keeps the original event attached, so the investigation starts with evidence.
- What it brings together
- The original alert and its timeline
- For your team
- Your team knows what happened and where to look
Automation with accountability
Move the work forward.
Keep the decision yours.
From a source alert to a recorded outcome, with a clear point for human review. Connector actions depend on your environment and policy.
- 01 / TRIGGER
Alert received
Source and event context
- 02 / INVESTIGATE
Enrich evidence
Tools and entity relationships
- 03 / APPROVAL
Awaiting review
Analyst evaluates the proposal
PAUSED - 04 / RESPONSE
Connector action
Only after required approval
NOT EXECUTED
More of the investigation.
Less of the legwork.
Help your analysts gather the context they need, understand related activity and prepare a finding they can explain. Connected services determine which checks are available.
What SOCPilots helps your team investigate
Related alerts
What happenedBring relevant security events together so an analyst can see the sequence, not just one isolated warning.
Reputation checks
What is knownCheck available threat-intelligence sources for information about an address or file. Reputation supports an investigation; it does not decide the outcome.
Earlier investigations
What came beforeFind similar cases and earlier decisions to give the team useful history.
Account activity
Who is involvedUnderstand relationships and unusual behaviour around the affected account.
Business context
What mattersBring device ownership and business importance into the investigation so the team can prioritize.
Internet exposure
What is visibleReview available information about externally exposed services as part of the wider picture.
Checks and observations stay with the case, so your team can review how a finding was reached.
Grounded in ATT&CK and in what you already decided
Retrieval sits underneath the agent. Technique descriptions, detection content and the SOC’s own written procedures are embedded and searched semantically, so an answer about lateral movement is anchored to the framework rather than improvised.
Uploaded evidence is indexed the same way. A document attached to a case last quarter is findable during an investigation this quarter, including text pulled out of screenshots.
- T1003.001LSASS memory0.89
- T1550Alternate material0.81
- proc-14Credential theft response0.77
- SP-1904Prior case, same host0.74
Automation that survives a restart
Playbooks are drawn on a canvas and stored as a graph. Runs are persisted, so a playbook waiting three days for an approval is still waiting after a deploy, and a step that was half-finished when a container restarted does not silently vanish.
Sixteen step types are available on the canvas: calling the agent, invoking a connector, asking a human to approve, branching on a condition, waiting, notifying, updating the case. Five trigger types start a run — a person, an incoming alert, a record event, a schedule, or a signed webhook.
Pause. Review. Resume.
Keep the execution record attached to the work, including the point where human approval is required.
View the response flowWhere the models come from
Your account, your spend
Model access runs on keys you supply. Self-hosted deployments keep that relationship entirely between you and the provider, and nothing is routed through us.
More than one provider, by design
Different work is routed to different models, and a provider outage falls through to another rather than taking investigation offline. Provider choice is configuration, not a rebuild.
You can see what it cost
Model calls are traced and priced, per case and per tenant, because “the AI feature” turning into an unexplained invoice is how these programmes get cancelled.
The AI parts are optional
Automated disposition and agent investigation can be turned off entirely and the platform still works as a SIEM-agnostic console. Nothing depends on a model being available.
Watch it investigate something real
The demo we would rather give you is one where the agent runs against your own alerts and you get to disagree with it.