Data Center Compliance and Security Agent using Google ADK and o3
India's Data Centre Security Bill mandates strict continuous compliance and automated audit logging across cloud workloads. Build an agent using Google ADK integrated with o3 reasoning models to parse security logs and enforce regulatory policies, achieving 95% breach indicator detection accuracy in test simulations.
What you are building
Build an automated cloud infrastructure audit agent using Google Agent Development Kit (ADK) paired with OpenAI o3 model reasoning for complex compliance detection.
How work is evaluated
Evaluates log audit reasoning, severity scoring, and compliance violation detection accuracy.
Datasets
Public datasets, plus the private datasets you have access to.
How submissions are scored
critical_threat_detection
Verifies that high-severity database dump anomalies are successfully flagged.
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
detection_accuracy
Accuracy of security anomaly classification. • target: 0.95 • range: 0-1
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
Learning goals
Build modular agents using Google Agent Development Kit (ADK)
Incorporate o3 reasoning model capabilities for complex policy deduction
Analyze cloud infrastructure syslog streams and access logs for security anomalies
Generate automated compliance failure reports aligned with national data centre standards
Resources
Synthetic dataset of 1,000 data center audit logs including simulated breach scenarios.
Challenge setup
Evaluates log audit reasoning, severity scoring, and compliance violation detection accuracy.
Challenge input
JSON array of cloud audit syslog events.
Google ADK
Provides modular agent component abstractions.
o3
Delivers complex logical reasoning across unstructured policy logs.
Evaluated output
JSON containing non_compliant_events count, critical_findings array, and overall_compliance_status.
- Verifies that high-severity database dump anomalies are successfully flagged.
- Accuracy of security anomaly classification. • target: 0.95 • range: 0-1
- Benchmark: Cloud Security Policy Reasoning Benchmark
- Detection Accuracy target: 0.95
- 1 public reference case
- Python execution harness
- Python sandbox (unavailable on Versalist)
Shows the first task, up to three tools, and up to three evidence items.
View technical recipe
Configured tools
- Google ADK · Required
- Voiceflow · Optional
- o3 · Optional
Evaluation contract
- critical_threat_detection · Weight 1
- detection_accuracy · Weight 1
Metrics
- detection_accuracy · Target 0.95
Benchmarks
- Cloud Security Policy Reasoning Benchmark
- security
Reference cases
- 1 public reference case
Workflow tasks
- Task 1
Input: JSON array of cloud audit syslog events.
Output: JSON containing non_compliant_events count, critical_findings array, and overall_compliance_status.
Execution harness: Python
Evaluation tools
- Google ADK · Agent Framework
- o3 · Reasoning Model
Recipe freeze marker
No recipe freeze time is recorded.