witfoo/precinct6-cybersecurity-100m

WitFoo Precinct6 Cybersecurity Dataset (large)

Version 2.0.0 (built 2026-09-11). Regenerated to address feedback from the University of Canterbury PIDS evaluation: attacks and normal traffic now share a timeline, usernames are wired into the provenance graph, mis-parsed timestamps are repaired, and every number in this card is generated from the uploaded data.

Overview

A large-scale, labeled cybersecurity dataset derived from production Security Operations Center (SOC) data processed by WitFoo Precinct version 6.x. It contains 114,421,340 sanitized security events captured live from 5 organizations (2024-07-26 11:10:20 UTC to 2024-08-28 00:36:03 UTC), 60,575 incident provenance graphs with their 238,511 embedded triggering signals, per-incident GraphML files, natural-language attack reports, and a merged provenance graph (67,172 nodes, 101,338,165 edges).

Available in two sizes (same incidents, same methodology, same sanitization registry):

• witfoo/precinct6-cybersecurity — 2,011,674 live signals, incidents in context
• witfoo/precinct6-cybersecurity-100m — the full live capture (this dataset)

Generate your own: WitFoo Precinct 6.x customers can create datasets from their own data with the open-source pipeline witfoo/dataset-from-precinct6.

This dataset supports research in:

• Provenance graph-based intrusion detection (KnowHow, NodLink, and similar systems)
• AI-driven cyber defense simulation (CybORG and MARL-based defense policy training)
• Security alert classification (malicious vs. suspicious vs. benign event labeling)
• Attack lifecycle analysis using MITRE ATT&CK framework mappings
• Detection rule evaluation using WitFoo's 261 lead detection rules

What changed in v2

Issue in v1	v2
Benign traffic and labeled attacks never overlapped in time (benign from 26 July 2024, incident leads ending 18 July 2024)	signals is the live capture labeled in place: 7,728 live rows are leads of 2739 incidents observed during the capture, surrounded by their ordinary traffic. Historical incident leads moved to a separate incident_signals table with an origin column so the two timelines cannot be confused.
Usernames existed on signals but credential nodes only inside incident objects	67,940,244 user → host edges (USER_ACTION) join CREDENTIAL nodes to the hosts they act on; incident cred nodes and live username fields share the same USER-NNNN tokens.
Handful of negative timestamps (Precinct date-parse failures)	Every lead timestamp is validated; 26 were repaired from the artifact timeuuid / rt= field / incident time, with the source recorded in timestamp_source. No timestamp in either table is outside 2000–2030.
README counts drifted from the uploaded files	This card is rendered from the build's metadata.json; signals/metadata.json and graph/metadata.json are uploaded alongside.
Incident nodes were keyed by Precinct uuids, disconnected from the live graph	Incident host/credential nodes use the same node ids as live signals (sanitized IP / user:USER-NNNN), so incident subgraphs overlay the live graph.
Aho-Corasick sweep occasionally corrupted uuids (id_raw) inside incidents	Record identifiers are protected from the sweep; leads can be joined to live rows on artifact_id.

Versions

This is v2.0.0, a full regeneration that replaces and withdraws the 2026-05 release.

The earlier release is not kept for download. Re-verifying it against this build's tooling showed that some values had escaped sanitization — device and account names survived inside JSON-escaped Windows event text and in stream_name — so it has been withdrawn rather than preserved at a tag. Anything built on it should be regenerated from v2.0.0.

Tokens are not comparable between the two releases in any case: each build maps identifiers with its own registry, so HOST-0042 in the old release is a different machine from HOST-0042 here.

from datasets import load_dataset

signals = load_dataset("witfoo/precinct6-cybersecurity-100m", "signals", split="train")   # v2.0.0

Quick Start

from datasets import load_dataset

REPO = "witfoo/precinct6-cybersecurity-100m"

# Live capture, labeled in place (benign / suspicious / malicious share hosts and hours)
signals = load_dataset(REPO, "signals", split="train")

# In-window attacks with their surrounding traffic: filter on the incident ids
attacks = signals.filter(lambda x: x["label_binary"] == "malicious")

# Historical incident leads (2022–2024) — a separate timeline, same columns
incident_signals = load_dataset(REPO, "incident_signals", split="train")

# Provenance graph: hosts + credentials, host->host event edges, user->host edges, incident links
nodes = load_dataset(REPO, "graph_nodes", split="train")
edges = load_dataset(REPO, "graph_edges", split="train")

# Deterministic attack reports (one per incident)
reports = load_dataset(REPO, "attack_reports", split="train")

# Full incident graphs (nested dicts keyed by uuid; not a typed config)
import pandas as pd
incidents = pd.read_json("hf://datasets/" + REPO + "/graph/incidents.jsonl", lines=True)

Join an incident lead to its live row: incident_signals.artifact_id == signals.artifact_id (both are the Precinct artifact timeuuid). Rows of signals that are leads carry the incident ids in incident_ids.

Temporal coverage

signals (live capture, origin = live) — 114,421,340 rows, 2024-07-26 11:10:20 UTC → 2024-08-28 00:36:03 UTC

Label	Rows	Share	Earliest	Latest
benign	113,798,478	99.46%	2024-07-26 11:10:20 UTC	2024-08-28 00:36:03 UTC
suspicious	615,134	0.54%	2024-07-26 11:10:26 UTC	2024-08-01 05:50:58 UTC
malicious	7,728	0.01%	2024-07-26 11:15:48 UTC	2024-07-30 04:56:35 UTC

incident_signals (embedded incident leads, origin = incident_lead) — 238,511 rows, 2022-05-30 14:43:45 UTC → 2024-12-31 23:58:09 UTC

Label	Rows	Share	Earliest	Latest
benign	0	0.00%	-	-
suspicious	0	0.00%	-	-
malicious	238,511	100.00%	2022-05-30 14:43:45 UTC	2024-12-31 23:58:09 UTC

The live capture is the complete artifact retention window of the archived Precinct cluster. Coverage is not uniform across organizations or days: check the per-organization table below, the per-label time ranges in signals/metadata.json, and the per-organization hourly histogram in build/label_stats.json before assuming a continuous capture.

Organizations

Organization	Live rows	Malicious (in place)	Suspicious
ORG-0004	99,556,527	4,176	196,942
ORG-0003	10,911,832	16	3,819
ORG-0005	2,594,196	3,536	411,433
ORG-0001	1,358,762	0	2,940
ORG-0002	23	0	0

Incidents in context

2739 incidents have at least one triggering signal that was found among the live rows; those 7,728 lead artifacts are labeled malicious in place in signals (they are not duplicated in incident_signals). Leads of 57,846 incidents were not found among the live rows (their incidents pre-date the capture, or the live artifact was not retained) and live in incident_signals.

Incident	Leads matched to live rows
acee85e0-4b4e-11ef-a07e-73bb772fb986	501
04bdda80-4b51-11ef-98d0-55d447741aef	445
ac8def50-4b4e-11ef-a07e-73bb772fb986	120
f4e66920-4df2-11ef-ab9d-d9f9158e7fcb	100
f4f2ec40-4df2-11ef-ab9d-d9f9158e7fcb	100
f5019240-4df2-11ef-ab9d-d9f9158e7fcb	100
f50dc740-4df2-11ef-ab9d-d9f9158e7fcb	100
f51ba9f0-4df2-11ef-ab9d-d9f9158e7fcb	100
f5282d10-4df2-11ef-ab9d-d9f9158e7fcb	100
f5365de0-4df2-11ef-ab9d-d9f9158e7fcb	100
f542e100-4df2-11ef-ab9d-d9f9158e7fcb	100
f551d520-4df2-11ef-ab9d-d9f9158e7fcb	100
f55ef480-4df2-11ef-ab9d-d9f9158e7fcb	100
d5e53480-4df1-11ef-b093-ed1dc3d0dbe9	100
d4ab54f0-4df1-11ef-8cc7-db879965e06f	100
...	...

Label distribution

Across both tables (114,659,851 rows):

Label	Rows	Share
benign	113,798,478	99.25%
suspicious	615,134	0.54%
malicious	246,239	0.21%

Disposition of malicious rows (raw Precinct incident status, see Ground Truth):

signals:

Disposition	Rows
Cold Case	3
Disrupted	5,779
Open	15
Unprocessed	1,931

incident_signals:

Disposition	Rows
Cold Case	8
Dismissed	5
Disrupted	176,510
Open	336
Unprocessed	61,652

Signal columns

Both signal tables share one schema (38 columns).

Column	Type	Description
timestamp	float	Unix epoch seconds. Live rows: Precinct ingest time (artifact timeuuid). Incident leads: lead observation time (repaired when invalid, see timestamp_source).
event_time	float	Event time reported by the source product (starttimeutc) when available; NaN otherwise.
timestamp_source	string	Where timestamp came from: artifact.timeuuid, lead.observed_at
origin	string	live (captured artifact) or incident_lead (artifact embedded in an incident)
org_id	string	Sanitized organization (ORG-NNNN)
artifact_id	string	Precinct artifact timeuuid — join key to graph/edges.jsonl (attrs.artifact_id). The two signal tables are disjoint on this key: a lead whose live row is in the capture appears in signals with label_binary = malicious, never in incident_signals
message_type	string	Event classification (e.g., firewall_action, account_logon, 4624, AssumeRole)
stream_name	string	Source product/data stream
pipeline	string	Ingestion pipeline
src_ip, dst_ip	string	Source/destination IP (sanitized)
src_port, dst_port	string	Ports
protocol	string	Network protocol (6=TCP, 17=UDP, 1=ICMP)
src_host, dst_host	string	Source/destination hostname (sanitized)
username	string	Associated account (USER-NNNN, sanitized; shared with incident credential nodes)
action	string	Event action (block, permit, logon, logoff)
severity	string	Severity level
vendor_code	string	Vendor-specific event code
message_sanitized	string	Full sanitized raw log message
label_binary	string	malicious, suspicious, or benign
label_confidence	float	Confidence in the tier (0.0–1.0). See Scoring.
attack_techniques	string	JSON array of MITRE ATT&CK technique IDs
attack_tactics	string	JSON array of MITRE ATT&CK tactic IDs (TA0001-style)
defense_techniques	string	JSON array of MITRE D3FEND technique IDs
mo_name	string	Modus operandi of the parent incident (e.g., Data Theft)
suspicion_score	float	Precinct incident suspicion score (0.0–1.0); 0 for benign/suspicious
lifecycle_stage	string	Kill-chain stage (initial-compromise, complete-mission, ...)
disposition	string	Raw Precinct incident status (Disrupted, Resolved, Dismissed, False Positive, Cold Case, Open, Unprocessed)
disposition_category	string	confirmed-malicious, false-positive, dismissed, automated
is_false_positive	bool	Analyst marked the parent incident a false positive
status_name	string	Same as disposition
incident_ids	string	JSON array of incident UUIDs (a lead can belong to several incidents)
matched_rules	string	JSON array of matched WitFoo lead rule descriptions
set_roles	string	JSON array of WitFoo classification roles (Exploiting Host, C2 Server, ...)
product_name, vendor_name	string	Security product and vendor

Graph data

Component	Count
Nodes	67,172 (HOST: 46,232, CREDENTIAL: 20,904, SERVICE: 6, FILE: 29, ACTOR: 1)
Edges	101,338,165
User → host edges (USER_ACTION)	67,940,244
Incident link edges	345,083
Incidents (per-incident GraphML)	60,575

• Node ids. Public IPs are global node ids; private IPs, hostnames and credentials are scoped by organization (ORG-0004/10.44.0.7, ORG-0004/user:USER-0007) because the same private address or account name exists in several customer networks. Incident host and credential nodes are mapped onto the same ids (their Precinct uuids are kept in attrs.precinct_node_ids). Every node carries the same attribute keys; first_seen/last_seen/signal_count Other incident node types (SERVICE, FILE, ACTOR, ...) keep their Precinct uuid as node id. come from live signals only, incident_first_observed/incident_last_observed from incident membership.
• Edges from signals carry the signal's labels, attrs.origin, attrs.org_id and attrs.artifact_id; edge type is derived from the message type (NETWORK_FLOW, LOGON, AUDIT_EVENT, ...). A signal with a username adds a USER_ACTION edge from the credential node to the accessed host (attrs.host_role says whether that was the destination or the reporting host).
• Edges from incidents are INCIDENT_LINK with the incident's labels; timestamp is the edge's own start time.
• Edge types: USER_ACTION (67,940,244), AUDIT_EVENT (31,864,472), NETWORK_FLOW (643,219), EVENT (511,133), INCIDENT_LINK (345,083), DNS_RESOLVE (34,014).
• The merged graph is shipped as edge shards only (too large for one GraphML); use the per-incident GraphML files or stream the edge files.

Attack reports

graph/attack_reports.jsonl holds one natural-language threat-hunting report per incident (60,575), deterministically composed from the incident's structured metadata (modus operandi, set roles, lead descriptions, MITRE mappings, timestamps). Each report states that it reflects Precinct's automated correlation output, not an independent investigation. Derivation: src/precinct6_dataset/attack_reports.py.

Files

• signals/signals-NNNNN.parquet — live signals (58 shards)
• signals/incident_signals.parquet — embedded incident leads
• signals/metadata.json — exact counts, per-label time ranges, org/stream/message-type distributions
• graph/nodes.jsonl, graph/edges-NNNNN.jsonl.gz (11 gzipped shards) — merged provenance graph (NDJSON)
• graph/incidents.jsonl — full sanitized incident records with embedded nodes, edges, leads (dicts keyed by Precinct uuids, so this file is not exposed as a load_dataset config; read it with pandas.read_json(lines=True))
• graph/incidents_graphml/<x>/<incident_id>.graphml — one GraphML per incident (sharded by first hex character)
• graph/attack_reports.jsonl — attack reports
• graph/metadata.json — graph counts and node id scheme
• reference/lead_rules_catalog.json — 261 lead detection rules, 158 products, 106 classification sets

Labeling methodology

Three-tier labels:

• malicious — the event is a triggering signal (lead) of a Precinct incident. In signals these are live rows joined to their incident on artifact_id; in incident_signals they are the embedded copies of leads whose live rows fall outside the capture.
• suspicious — the event matched one or more of WitFoo's 261 lead detection rules but is not a lead of any incident.
• benign — no rule matched and the event is not part of any incident.

A lead that belongs to several incidents is one row whose incident_ids lists them all; its mo_name, disposition and suspicion_score come from the highest-suspicion incident.

Ground truth and disposition

All labels derive from WitFoo Precinct's automated incident correlation engine — there is no independent, analyst-verified ground truth. Treat Precinct as a strong but imperfect oracle and use disposition to stratify by analyst review:

disposition	Meaning	Confidence in label
Disrupted	SOC analyst confirmed the incident and intervened	High
Resolved / Confirmed / Investigating	Analyst confirmed / engaged	High–Medium
Open	Queued for an analyst, no decision yet	Lower
Cold Case	Aged out without an analyst decision	Lower
Dismissed	Analyst dismissed the incident	Negative
False Positive	Analyst confirmed false positive	Negative
Unprocessed	Automated detection only	Lower

disposition_category buckets these into confirmed-malicious, false-positive, dismissed and automated.

Scoring

• suspicion_score — Precinct's proprietary score of the parent incident (0–1). Zero for benign and suspicious.

• label_confidence — how much corroborating evidence supports the tier (not a probability of maliciousness):

  Label	Formula
  malicious	max(0.6, suspicion_score) clamped to 0.95; 0.3 if is_false_positive
  suspicious	0.4 + 0.1 × n_matched_rules + 0.05 × n_set_roles, clamped to [0.5, 0.85]
  benign	0.5

MITRE ATT&CK mappings

Tactics and techniques are derived from (1) WitFoo set role names on the incident, (2) the incident's modus operandi, and (3) per-product framework data embedded in incident.nodes.products.frameworks, deduplicated. They are priors, not analyst-confirmed per-event attributions. Mapping tables: src/precinct6_dataset/mitre_mapping.py.

Source products

The events in this build come from 20 security products across 13 vendors (exact counts in signals/metadata.json under product_distribution / vendor_distribution).

Most frequent products: AWS Instance Backup, Windows Logs, Windows Active Directory, VMWare VCenter, ASA Firewall, Linux PAM, AWS VPC Security, Barracuda WAF, Cisco Network Operating System, Symantec Endpoint Protection, ManageEngine ADManager, Falcon, Duo, Precinct. Vendors: Amazon Web Services, Microsoft, VMWare, Cisco, Linux, Barracuda, Symantec, ManageEngine, Crowdstrike, WitFoo, Apache, SentinelOne.

The generator's rule catalog (reference/lead_rules_catalog.json) covers a much wider set — 158 products across firewalls, endpoint protection, network detection, identity, cloud, email security and infrastructure — because it is shared by every deployment; only the products above actually appear in this capture.

Top streams in this build: aws_cloudtrail_events (55,058,698), windows_security_audit (31,864,472), microsoft-windows-security-auditing (11,004,179), aws_cloud_trail (6,224,538), no_useful_info (4,698,096), vcenter (1,203,516), java_stack_trace (696,451), mssql$HOST-1053 (657,584), crond (435,510), cisco_asa (415,269).

Sanitization

All customer-identifying information was removed with the open-source four-layer pipeline (witfoo/dataset-from-precinct6):

1. Structured field sanitization + Aho-Corasick multi-pattern sweep — deterministic tokens (public IPs → RFC 5737 TEST-NET, private IPs → HMAC-remapped RFC 1918, hostnames → HOST-NNNN, accounts → USER-NNNN, organizations → ORG-NNNN, emails → user-NNNN@example.net, SIDs, AWS accounts/ARNs, machine accounts), then a sweep over every string field with the full registry. Record identifiers (artifact/incident uuids) are protected from the sweep.
2. Format-specific log parsing — Cisco ASA, Windows Security XML, WinLogBeat, AWS CloudTrail, Palo Alto, VMware vCenter, DNS, and a generic fallback.
3. ML residual detection — Microsoft Presidio (spaCy) and BERT NER on a stratified sample; findings trigger full re-sanitization.
4. LLM contextual review — sampled review by the local WitQ model (ran for this build: 1,500 records reviewed by witq, 3 additional values registered). The ML layer (3) sampled 3,000 records and registered 452 additional values.

The same original value always maps to the same token across both signal tables, the incidents and the graph, so topology and identity are preserved. Both dataset sizes were produced from one registry, so tokens agree between them. The registry for this build holds 85,079 mappings.

Research context

Produced in collaboration with the University of Canterbury (New Zealand) Computer Science and Software Engineering department for two research projects: an AI cyber-security battle simulator (improving CybORG with realistic IDS observations and graph-based defense policies) and intrusion detection based on provenance graphs (evaluating KnowHow, NodLink and similar PIDS).

Limitations

• Label imbalance reflects production SOC reality; sample accordingly.
• Temporal scope: the live capture spans 2024-07-26 11:10:20 UTC to 2024-08-28 00:36:03 UTC with uneven per-organization coverage; incident leads span 2022-05-30 14:43:45 UTC to 2024-12-31 23:58:09 UTC.
• Ground truth: labels are Precinct's automated correlation; stratify with disposition.
• Sanitization trade-offs: some free-text detail is reduced by PII replacement.
• Tokens differ from v1: the v2 registry was rebuilt, so USER-NNNN / HOST-NNNN / ORG-NNNN values do not correspond to v1 values.

Citation

@dataset{witfoo_precinct6_100m_2026,
  title={WitFoo Precinct6 Cybersecurity Dataset (large)},
  author={WitFoo, Inc.},
  year={2026},
  version={2.0.0},
  url={https://huggingface.co/datasets/witfoo/precinct6-cybersecurity-100m},
  license={Apache-2.0}
}

License

Apache License 2.0