How it works overview
Phase 01
Plan First
Analyze before you infer.
Overview
Every Devtor session starts with structured planning. Instead of sending your prompt straight to the most capable model, the orchestrator breaks down what you are asking for, what files matter, and what level of reasoning the task actually requires.
In depth
Plan-first mode surfaces complexity scores, token estimates, and a recommended model tier before any inference runs. You see projected cost and can approve, adjust context, or cancel — so expensive models are never invoked blindly on work a cheaper agent could handle.
What happens
- 1Parse the natural-language task and intent
- 2Score complexity (low / medium / high)
- 3Estimate tokens and projected USD cost
- 4Select candidate models and context files
- 5Present the plan for approval or adjustment
Terminal output
terminal
$devtor run "Refactor auth module"
Analyzing task complexity...
→ Complexity: medium
→ Estimated tokens: 2,400
→ Optimal model: claude-3-5-haiku
→ Est. cost: $0.006
Benefits
- Transparent cost estimates before execution
- Task complexity scoring drives better routing
- Context attachments chosen for relevance, not volume
- Plans you can review and redirect mid-flow
Related capabilities
Ready to orchestrate?
Installation guide and CLI — coming soon.