Structured AI decision support through multiple independent perspectives
The orchestration engine for coding agents
Stage
Not given
MVP
Founded
Not given
Not given
Based in
Not given
Not given
Pricing model
Paid
Free trial
Pricing
Essential: €9 (4 AI perspectives); Consilium: €19 (8 AI perspectives); Grand Consilium: €29 (12 AI perspectives). Prepaid packages available: Essential packages of 5 for €39 or 10 for €69; Consilium packages of 5 for €69 or 10 for €129; Grand Consilium packages of 5 for €99 or 10 for €179. Each pric
$12/month or $96/year (save 33%), 800 credits per month, one credit per task. New accounts get 800 free credits for 30 days with no card required.
Key features
Multiple independent AI perspectives analyze the same question
Anonymous cross-checking of AI answers
Structured Decision Brief with recommendation and risks
Three analysis levels: Essential (4), Consilium (8), Grand Consilium (12) perspectives
Support for 42 languages
Automatic masking of personal and identifying data
Zero Data Retention requirement from AI providers
One follow-up question included per report
Declare tasks and dependencies in a workflow file
Deterministic task scheduling in topological order
Fresh session per task with isolated context
Watch tasks live, pause, kill, or recover individual tasks
Automatic merge conflict resolution by agents
Results collected in integration branch with main untouched
Works with Claude Code and Codex agents
Runs locally on your machine
Code never uploaded to CodeDAG servers
What makes it different
Anonymous ranking ensures quality of arguments over model reputation
Structured decision report rather than raw AI text
Data confidentiality with Zero Data Retention built in
Fixed transparent pricing per analysis level
No babysitting required; autonomous agents run the entire workflow
Runs entirely on your machine with no code uploads
Billed per task at one credit each, regardless of duration or complexity
Works with your existing agent subscriptions without additional token charges
Full control: watch, review, and manually steer any task