The FTL2 roadmap outlines planned features across three time horizons: near-term (Terraform provider integration, multi-cloud validation, module discovery API, policy engine improvements), medium-term (declarative resource planning, Ansible-to-FTL2 converter, transaction/rollback), and longer-term (event-driven workflows, desired state as markdown). The roadmap reveals FTL2's strategic direction toward becoming a unified automation platform that bridges Ansible modules, Terraform providers, and AI-driven infrastructure management.
Key Concepts
Terraform Provider Integration — access 3000+ Terraform providers via gRPC protocol bridge (Python to Go), with state mapping to .ftl2-state.json and plan/apply semantics
Module Discovery API — runtime module discovery (ftl.list_modules(), ftl.describe()) designed specifically for AI-generated scripts to select correct modules
Declarative Resource Planning — Terraform-style resource declarations with automatic dependency inference, topological sort for parallel execution waves, and reverse-order teardown
Transaction/Rollback — atomic operations across multiple modules with explicit rollback on failure and savepoint/restore patterns
Desired State as Markdown — natural-language infrastructure specs where AI generates disposable scripts, state file is ground truth, and drift detection compares spec against actual state
Policy validation on load (warn about unknown match keys)
Policy composition (multiple files, directory of policies)
Policy summary surfaced in audit log
Relationships
Terraform providers extend the module ecosystem beyond Ansible collections, connecting to the existing module calling patterns and state management system
Module Discovery API connects to the AI-loop rules system — AI agents need runtime introspection to select modules
Multi-cloud validation proves the automation context API (inventory, group proxies) is cloud-agnostic
Declarative resource planning builds on the existing .ftl2-state.json state tracking and idempotency model
Ansible-to-FTL2 converter directly relates to migration from Ansible — formalizes the mapping that users currently do manually
Transaction/rollback extends the existing crash recovery and state file mechanisms
Event-driven workflows builds on the gate process architecture (gate modules emitting events back to controller)
Desired state as markdown connects state files, drift detection, and AI-generated scripts into a single workflow
Exam-Relevant Points
Terraform integration requires a gRPC protocol bridge from Python to Go — FTL2 is Python, Terraform providers are Go
State from Terraform providers maps to the same .ftl2-state.json format used by native FTL2 state tracking
The Ansible-to-FTL2 conversion mappings are specific: with_items→for loops, when→if statements, handlers→explicit service restarts, roles→function calls
Declarative resource planning uses topological sort for parallel execution and reverse order for teardown
Module Discovery API has two functions: list_modules (browse by category) and describe (detailed parameter info)
In the desired-state-as-markdown model, scripts are disposable/regenerated — the state file is ground truth, not the scripts
Policy engine improvements include composition (multiple files) and validation on load, but short-circuiting means rules can do more than just deny