Clioloop's Agentic Fusion is not just "run the same prompt 5 times and pick the best." It's a structured pipeline where different models play different roles, with strict security boundaries between them.
When you run /fusion
, up to 5 planner models are dispatched in parallel. Each planner:
The planners might suggest different strategies:
Your main model takes the planners' proposals and does the actual work:
This is the key difference from "ensemble" approaches: the main model does real work with real tools, not just text generation.
Up to 5 reviewer models critique the draft:
The draft is revised based on reviewer feedback:
Everything combines into one answer that has already passed independent review.
The safety comes from schema-level restrictions:
| Role | Can Read | Can Write | Can Execute |
|---|---|---|---|
| Planners | ✅ Files, web, images | ❌ Nothing | ❌ Nothing |
| Main Model | ✅ Everything | ✅ Files | ✅ Commands |
| Reviewers | ✅ Draft, files, images | ❌ Nothing | ❌ Nothing |
Planners and reviewers are read-only by construction. They can research and critique, but they can never touch your files or run commands. Only your main model has tool access, and you watch it work live.
The quality comes from synthesis, not brute force. Cheap open models combine into something that rivals a frontier model:
You get frontier-level quality at a fraction of the cost, because you're paying for cheap models to plan and review, and only one model does the expensive tool work.
curl -fsSL https://raw.githubusercontent.com/Clioloop/Clioloop-agent/main/scripts/install.sh | bash
clio setup
clio
Then run /fusion
on any task.
Feedback welcome!