16 exercises match.
LAB-280-02 · Intermediate · prototype
Intermediate Repository Change Lab
Review a fixed two-file proposal, enforce source hashes and an explicit write allowlist, reject stale approval, and run separate Python tests. This deterministic harness does not invoke an LLM or isolate the host.
- Hardware
- Laptop / CPU, Python 3.10+; no model download. Docker optional.
- Environment
- python-stdlib, Docker optional
- Availability and validation
- Python-only validation recorded (11 safety checks, one symlink check skipped); Docker and independent reproduction not performed. Recorded automated run: 2026-10-21; independent reproduction: not-started
Open guideLAB-280-00 · Beginner · prototype
Basic Local Coding Agent Lab
Preview a single-file Python correction with local Ollama inference, approve a fresh proposal, and run separate Docker tests with networking disabled. Host isolation and signed audit evidence are outside this prototype.
- Hardware
- Host capable of running qwen2.5-coder:7b; memory and latency depend on quantization and hardware. Docker required for container tests.
- Environment
- Python 3.10+, Ollama / qwen2.5-coder:7b, Docker Compose / Linux containers
- Availability and validation
- Public introductory prototype; target-host run evidence is not published. No automated validation record published
Open guideLAB-100-01 · Beginner · reviewed
Sovereignty Boundary Mapping
Create a system definition, trust-boundary map, control objectives, and evidence plan.
- Hardware
- See exercise requirements; capacity not benchmarked
- Environment
- Architecture/design workspace
- Availability and validation
- Design exercise; implementation pending. No automated validation record published
View module and planned exerciseLAB-110-01 · Beginner · reviewed
Private AI Threat Model
Develop and prioritize abuse and failure scenarios for a private AI workload.
- Hardware
- See exercise requirements; capacity not benchmarked
- Environment
- Architecture/design workspace
- Availability and validation
- Design exercise; implementation pending. No automated validation record published
View module and planned exerciseLAB-120-01 · Beginner · reviewed
Sovereign Reference Architecture
Produce and defend a multi-view reference architecture and decision set.
- Hardware
- See exercise requirements; capacity not benchmarked
- Environment
- Architecture/design workspace
- Availability and validation
- Design exercise; implementation pending. No automated validation record published
View module and planned exerciseLAB-200-01 · Advanced · draft
Controlled Artifact Promotion
Design artifact quarantine, approval, promotion, rollback, and recovery evidence.
- Hardware
- See exercise requirements; capacity not benchmarked
- Environment
- Architecture/design workspace
- Availability and validation
- Design exercise; implementation pending. No automated validation record published
View module and planned exerciseLAB-210-01 · Advanced · draft
Model Lifecycle Decision
Evaluate and package a model candidate with scoped lifecycle approval.
- Hardware
- See exercise requirements; capacity not benchmarked
- Environment
- Architecture/design workspace
- Availability and validation
- Design exercise; implementation pending. No automated validation record published
View module and planned exerciseLAB-220-01 · Advanced · draft
Model Serving Decision
Measure a runtime profile and design governed routing, capacity, and recovery.
- Hardware
- See exercise requirements; capacity not benchmarked
- Environment
- Architecture/design workspace
- Availability and validation
- Design exercise; implementation pending. No automated validation record published
View module and planned exerciseLAB-230-01 · Advanced · prototype
Permission-Aware Private RAG Lab
Run a synthetic local retrieval system that enforces document permissions, citations, deletion, no-answer behavior, and injection controls.
- Hardware
- Laptop / CPU; no model download required
- Environment
- python-stdlib, docker-compose
- Availability and validation
- Runnable synthetic prototype with recorded validation. Recorded automated run: 2026-09-07; independent reproduction: not-started
Open guideLAB-240-01 · Advanced · draft
AI Gateway Policy
Test identity-aware allow, deny, redact, route, egress, and approval decisions.
- Hardware
- See exercise requirements; capacity not benchmarked
- Environment
- Architecture/design workspace
- Availability and validation
- Design exercise; implementation pending. No automated validation record published
View module and planned exerciseLAB-250-01 · Advanced · prototype
AI Release Evidence Pack
Assemble a control-to-evidence traceability matrix, evaluate evidence quality and exceptions, and issue a deterministic release recommendation for a synthetic system passport.
- Hardware
- Laptop / CPU; no model download required
- Environment
- python-stdlib
- Availability and validation
- Runnable synthetic prototype with recorded validation. Recorded automated run: 2026-09-07; independent reproduction: not-started
Open guideLAB-260-01 · Advanced · draft
AI Operations Scorecard
Create service objectives, incident evidence, capacity measures, and unit economics.
- Hardware
- See exercise requirements; capacity not benchmarked
- Environment
- Architecture/design workspace
- Availability and validation
- Design exercise; implementation pending. No automated validation record published
View module and planned exerciseLAB-270-01 · Advanced · draft
Restricted Cluster Recovery
Plan and validate restricted workload controls, disconnected updates, and recovery.
- Hardware
- See exercise requirements; capacity not benchmarked
- Environment
- Architecture/design workspace
- Availability and validation
- Design exercise; implementation pending. No automated validation record published
View module and planned exerciseLAB-280-01 · Advanced · outline
Governed Agent Design
Design bounded agent tools, approvals, confinement, memory, and action receipts.
- Hardware
- See exercise requirements; capacity not benchmarked
- Environment
- Architecture/design workspace
- Availability and validation
- Design exercise; implementation pending. No automated validation record published
View module and planned exerciseLAB-ARCH-01 · Advanced · draft
Architecture Control Room
Integrate system, supply-chain, runtime, knowledge, gateway, operations, and evidence decisions.
- Hardware
- See exercise requirements; capacity not benchmarked
- Environment
- Architecture/design workspace
- Availability and validation
- Design exercise; implementation pending. No automated validation record published
View module and planned exerciseLAB-AIRGAP-01 · Advanced · outline
Disconnected Platform Pilot
Promote approved artifacts into a simulated disconnected runtime and perform a controlled update and recovery.
- Hardware
- See exercise requirements; capacity not benchmarked
- Environment
- Architecture/design workspace
- Availability and validation
- Design exercise; implementation pending. No automated validation record published
View module and planned exercise