Review and Approve Actions
How the approval queue and confidence scoring work in Rills
The approval queue is where you review AI-proposed actions before they execute. This is the core of Rills' value: AI handles the monitoring and analysis, you make the final decision, and the system learns which decisions to escalate and which to handle autonomously.
Why approvals matter
Most automation tools operate on rigid if/then rules or run fully autonomously. The first option is inflexible, the second creates anxiety about what the AI might do unsupervised.
Rills takes a different approach: AI agents analyze data and propose actions, you approve what makes sense, and over time the system learns your preferences and handles routine decisions automatically while escalating edge cases.
This eliminates "monitoring anxiety" — you're never wondering what your automation is doing or worrying that it made a bad decision. You have the final word without constant manual oversight.
The approval flow
Here's how a typical approval works from trigger to execution:
1. Trigger fires
Your workflow trigger activates — a new email arrives, a scheduled time hits, a webhook delivers data. The workflow starts running.
2. AI agent analyzes
The AI agent step receives the trigger data and analyzes it according to your prompt. It considers context, looks for patterns, and decides what action to recommend.
3. Proposal appears in queue
The proposed action appears in your approval queue with:
- Context — What triggered the workflow and what data the AI analyzed
- Proposed action — What the AI wants to do (send this email, create this task, update this record)
- Reasoning — Why the AI recommends this action
- Confidence score — How certain the AI is about this recommendation
4. You review and decide
From your approval queue (accessible via web dashboard or the mobile-optimized web interface), you review the proposal and choose:
- Approve — Execute the action as proposed
- Reject — Don't execute, and record that this type of proposal should be rejected
- Edit — Modify the action before execution (change email text, adjust task details)
5. Approved actions execute
When you approve, the workflow continues and executes the action. The result is logged, and the workflow completes.
When you reject or edit, Rills records your decision and uses it to improve future confidence scoring.
Confidence scoring
Every proposed action has a confidence score from 0-100. This score represents how certain the AI is that you'll approve this action based on your past approvals and the current context.
How confidence works:
- Low confidence (0-40): The AI isn't sure what you'd want — edge case, ambiguous situation, or not enough past data. These always require manual approval.
- Medium confidence (40-70): The AI has seen similar situations and thinks it knows what you'd approve, but there's enough variation that it escalates for review.
- High confidence (70-100): The AI is very certain based on your approval history. You can optionally set auto-approval thresholds for these.
Confidence improves over time:
As you approve and reject proposals, the AI learns your preferences. A workflow that starts at 20% confidence (always escalates) can graduate to 80% confidence (rarely needs approval) after a few weeks of feedback.
This is how workflows learn — they start supervised, learn from your decisions, and gradually become autonomous for routine cases while still escalating unusual situations.
Auto-approval thresholds
Once a workflow has enough approval history, you can set an auto-approval threshold. For example:
- Auto-approve actions with confidence ≥ 85%
- Always require approval for confidence < 85%
This lets you gradually hand off routine decisions while maintaining control over edge cases. You can adjust confidence thresholds per AI node based on risk tolerance -- different steps in the same workflow can have different thresholds.
Mobile approval
The approval queue is accessible from your phone. The mobile-optimized swipe interface lets you approve or reject actions from your phone while commuting, between meetings, or from anywhere.
Approving on mobile takes seconds — swipe to see context, swipe to approve or reject. Your workflows keep running without requiring you to be at your desk.
Approvals are free
An important note about pricing: approvals are completely free. You're never charged for reviewing and approving actions, no matter how many proposals you review.
Rills encourages you to add approval steps wherever you want them, without worrying about costs. You only pay for AI agent calls (AI Credits) and external actions that execute (Workflow Credits), not for the human review process.
Next steps
You now understand how approvals work and how workflows learn from your decisions. Let's wrap up with how workflow credits work so you know exactly what you're paying for.