Multi-agent forensic intelligence

Investigate incidents at machine speed

ForensicAI turns raw artefacts into a defensible investigation: nine specialist agents parse your evidence, correlate the signals, score the threat and hand you a report you can put in front of a court.

9

Specialist agents

<3 min

Typical pipeline

0–100

Threat scoring

MITRE

ATT&CK mapped

Platform

Everything a forensic team needs, in one console

Automated pipelines

Upload evidence and the orchestrator dispatches every downstream agent in sequence, with full audit trails.

Threat intelligence

IOCs are enriched and mapped to MITRE ATT&CK techniques so findings land in a language your SOC already speaks.

Quantified risk

Every case gets a 0–100 threat score, a risk band and per-finding confidence — no unexplained verdicts.

Court-ready reports

Executive summary, technical detail, timeline, attack path and limitations, exportable for legal review.

Locked-down access

Administrator-approved accounts, role-based permissions, private evidence storage and immutable audit logs.

Explainable agents

Each agent publishes its reasoning, evidence references and recommendations alongside its findings.

Multi-agent system

Nine agents, one chain of custody

Each agent owns a narrow slice of the investigation and reports its own confidence, so you always know which conclusions to trust.

Orchestrator Agent

Mission control

Plans the investigation, decides which specialist agents to dispatch, and arbitrates conflicting findings.

LLM plannerTask graphStructured output schema

Evidence Collection Agent

Acquisition & chain of custody

Normalises uploaded artifacts, records hashes, and maintains an auditable chain of custody.

SHA-256MIME sniffingObject storage

Log Analysis Agent

Telemetry parsing

Parses Windows Event Logs, syslog, web and application logs to surface anomalous authentication and execution activity.

Windows Event Log parserRegex extractionStatistical baselining

Malware Detection Agent

Signature & heuristic scanning

Screens artifacts for known malicious signatures and suspicious structural indicators.

ClamAV signature engineHeuristicsEntropy analysis

Timeline Reconstruction Agent

Temporal ordering

Fuses timestamps across all sources into a single defensible incident timeline.

Temporal fusionClock-skew correctionLLM synthesis

Correlation Agent

Entity linking

Links accounts, hosts, files and network endpoints into an attack graph across independent evidence sources.

Graph correlationEntity resolutionLLM reasoning

Risk Assessment Agent

Impact scoring

Scores likelihood and impact to produce a defensible risk rating with the reasoning that produced it.

Weighted scoring modelLLM justification

Threat Intelligence Agent

Adversary context

Maps observed behaviour to known adversary techniques and provides contextual intelligence.

MITRE ATT&CKIOC extractionLLM knowledge

Report Generation Agent

Documentation

Compiles the investigation into an executive and technical report suitable for legal and management review.

LLM synthesisStructured report schemaPDF/HTML export

Workflow

From artefact to court-ready report

  1. 01

    Create the case

    Describe the incident and classify the investigation type.

  2. 02

    Upload evidence

    Logs, memory dumps, disk artefacts and network captures land in private storage.

  3. 03

    Run the pipeline

    The orchestrator dispatches all nine agents and streams progress.

  4. 04

    Review findings

    Inspect per-agent reasoning, timelines, IOCs and MITRE mappings.

  5. 05

    Export the report

    Publish an executive and technical report for legal review.

Bring your next investigation online

Request an analyst account. An administrator reviews every request before console access is granted.

Request access