Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,54 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Redteam-Operations-Datasets
|
| 2 |
+
|
| 3 |
+
A comprehensive dataset for training, benchmarking, and documenting offensive/red-team operations in Active Directory and hybrid environments.
|
| 4 |
+
|
| 5 |
+
## Structure
|
| 6 |
+
|
| 7 |
+
- `datasets/` — All dataset files, organized by attack phase, tactic, and technique.
|
| 8 |
+
- `datasets/schema/record.schema.json` — Canonical schema for dataset records.
|
| 9 |
+
- `datasets/*.jsonl` — Task-specific datasets, each line is a JSON object.
|
| 10 |
+
|
| 11 |
+
### Dataset Files
|
| 12 |
+
|
| 13 |
+
- `discovery.jsonl` — Host/service/AD enumeration.
|
| 14 |
+
- `initial_access.jsonl` — Phishing, exploitation, rogue device.
|
| 15 |
+
- `privilege_escalation.jsonl` — Escalation, ACL/service attacks.
|
| 16 |
+
- `credential_access.jsonl` — Dumping, cracking, relay, credential stores.
|
| 17 |
+
- `lateral_movement.jsonl` — SMB, WinRM, Kerberos, cloud movement.
|
| 18 |
+
- `persistence.jsonl` — Golden/silver tickets, scheduled tasks, backdoors.
|
| 19 |
+
- `collection_exfil.jsonl` — Data staging, compression, exfiltration.
|
| 20 |
+
- `c2.jsonl` — C2 infra, beacons, tunneling.
|
| 21 |
+
- `impact.jsonl` — Ransomware, destruction, shadow copy removal.
|
| 22 |
+
- `evasion.jsonl` — AV/EDR bypass, obfuscation, log tampering.
|
| 23 |
+
- `cloud_hybrid.jsonl` — Azure, Entra ID, O365, hybrid pivots.
|
| 24 |
+
- `tool_glossary.jsonl` — Tool descriptions, usage scenarios, command syntax.
|
| 25 |
+
- `chain_operations_examples.jsonl` — Multi-step attack chains/playbooks.
|
| 26 |
+
- `error_handling.jsonl` — Failure modes, fallbacks.
|
| 27 |
+
- `advanced_techniques.jsonl` — Latest CVEs, novel chaining, new exploits.
|
| 28 |
+
- `decision_matrix.jsonl` — When to use which technique; risk/noise comparisons.
|
| 29 |
+
|
| 30 |
+
## Schema
|
| 31 |
+
|
| 32 |
+
See [`datasets/schema/record.schema.json`](schema/record.schema.json) for full field list (instruction, input, output, meta).
|
| 33 |
+
|
| 34 |
+
## Usage
|
| 35 |
+
|
| 36 |
+
- For fine-tuning LLMs, benchmarking detection, or creating mindmaps/playbooks.
|
| 37 |
+
- Each file is a JSONL (one record per line), see the example schema below.
|
| 38 |
+
|
| 39 |
+
## Example Record
|
| 40 |
+
|
| 41 |
+
```json
|
| 42 |
+
{
|
| 43 |
+
"instruction": "Describe lateral movement using PsExec.",
|
| 44 |
+
"input": "psexec.py <domain>/<user>:<password>@<target_ip>",
|
| 45 |
+
"output": "Executes commands on remote host via SMB using provided credentials.",
|
| 46 |
+
"meta": {
|
| 47 |
+
"phase": "lateral_movement",
|
| 48 |
+
"required_privilege": "admin on target",
|
| 49 |
+
"risk_level": "high",
|
| 50 |
+
"references": ["https://github.com/fortra/impacket"],
|
| 51 |
+
"year": "2024"
|
| 52 |
+
}
|
| 53 |
+
}
|
| 54 |
+
```
|