| name | data-platform-attack |
| description | Attacks against cloud data platforms and analytics pipelines — Snowflake, Databricks, BigQuery, Redshift, dbt, Apache Airflow, and lakehouse architectures. Covers identity-based breaches (no perimeter), warehouse SQL injection at scale, IAM privilege escalation, secrets in DAGs, notebook code injection, and cross-tenant data exfiltration. Distinct from database-attack (protocol-level RDBMS) and cloud-security (broader CSP control plane). |
| origin | openclaw |
| version | 0.2.0.2 |
| compatibility | ["openclaw","claude-code","cursor","windsurf"] |
| allowed-tools | ["Bash","Read","Write","Edit","WebSearch","WebFetch"] |
| metadata | {"domain":"data-platform","tool_count":14,"guide_count":2,"mitre":"TA0006-Credential Access, T1078-Valid Accounts, T1213-Data from Information Repositories","last_reviewed":"2026-07-26"} |
Data Platform Attack
Supplementary Files:
payloads.md — Attack payloads and commands organized by platform: Snowflake, Databricks, BigQuery, Redshift, dbt, Airflow, Spark, and lakehouse cross-platform attackstest-cases.md — 18 structured test cases covering reconnaissance, identity abuse, SQL injection at warehouse scale, secrets harvesting, notebook injection, and exfiltrationguides/data-platform-attack-playbook.md — End-to-end playbook with engagement scoping, lab setup, reconnaissance methodology, identity abuse workflows, SQLi-at-scale patterns, exfiltration tactics, and blue-team detection engineeringguides/real-world-incident-case-studies.md — Eight case studies including the 2024 Snowflake/Lapsus$/Scattered Spider breach, Mandiant UNC5537, Databricks notebook abuse, Airflow DAG secret harvesting, and dbt CI token theft
Summary
This skill targets modern cloud data platforms — warehouses, lakehouses, orchestration layers, and transformation pipelines — where the boundary between analytics and production has collapsed. Attackers no longer tunnel through database listeners; they walk in through SSO, service tokens, and CI/CD pipelines.
Tools: snowsql, databricks-cli, gcloud (BigQuery), aws cli (Redshift/Athena), bq, psql (Redshift), dbt, airflow-cli, spark-sql, sqlmap, pacu, enumerate-iam, m365 dataset scrapers
Domain: data-platform
MITRE ATT&CK: TA0006 Credential Access · T1078 Valid Accounts · T1213 Data from Information Repositories · T1552 Unsecured Credentials
Description
Attacks against cloud data platforms and analytics pipelines — Snowflake, Databricks, BigQuery, Redshift, dbt, Apache Airflow, and lakehouse architectures. These platforms concentrate an organization's most valuable data behind identity boundaries rather than network boundaries, so the primary attack surface is identity abuse, misconfigured public endpoints, secrets embedded in pipelines, and SQL injection scaled across warehouses that contain every production table.
This skill is distinct from:
- database-attack — which targets RDBMS at the protocol/listener level (Oracle TNS, MySQL socket, Redis/MongoDB unauth) behind a network perimeter
- cloud-security — which targets CSP control planes (EC2/IAM/S3) broadly
- container-security — which targets OCI/K8s
Where database-attack asks "can I reach the listener?", data-platform-attack asks "given that the warehouse is internet-reachable by design and uses SSO/SAML, how do I turn a phished engineer into 4TB of exfiltrated customer records?"
