# Intent Layer Refinement Workflow
> [!summary]
> Use this workflow when Eval Labs reveals that Lucia is routing prompts into the wrong behavior mode.
---
## What an intent-layer failure looks like
An intent-layer failure happens when Lucia understands the words too shallowly and chooses the wrong mode.
Example:
```text
User: I feel totally out of the loop.
Bad response: I can help with operator priorities, arrivals, concierge readiness...
```
The user is not asking for a capability menu.
The user is expressing disorientation.
---
## Step 1 — Create a focused suite
Use 5–10 related prompts.
Example suite:
```text
I'm generally frazzled.
I feel totally out of the loop.
Ugh, I have no idea what to do.
I am so lost.
Kids just got home and I'm overwhelmed.
I'm frustrated and behind.
I don't trust that I know what's going on.
```
---
## Step 2 — Confirm the Eval Labs endpoint
Eval Labs runs against the Lucia endpoint configured in its environment.
For current Lucia v0.1.3.6 validation, the active target should be:
```text
https://api-dev.hellolucia.ai/admin/operator-focus
```
Staging should be used only when the validation pass has explicitly been configured or promoted for staging validation.
Do not let an example env file override the intended validation target for the pass.
---
## Step 3 — Review mode accuracy
For each response, ask:
```text
Did Lucia choose the right behavior mode?
```
If the response gives a generic redirect to a distress signal, fail it.
---
## Step 4 — Patch the correct owner
Likely owner:
```text
operatorFocusBrain.js
```
Possibly later:
```text
refineOperatorFocusOutput.js
```
Do not patch wording if the mode is wrong.
---
## Step 5 — Deploy to the configured target
Local-only changes are not enough for Eval Labs.
For current Lucia v0.1.3.6 validation, the Engine behavior must be deployed to dev before Eval Labs can test it.
If the pass is intentionally configured for staging validation, deploy or promote to staging as part of that pass and record that choice in the review thread.
---
## Step 6 — Re-run the exact same suite
Do not change prompts yet.
Run the same suite again.
Compare:
- which prompts now route correctly
- which still fail
- whether any neighboring behavior regressed
- whether tone improved or became too mushy
---
## Step 7 — Export evidence
Export the run after saving reviews.
Share the export with the engineering/product thread.
---
## Success standard
The patch is successful only if the repeated suite shows improved behavior without creating new failures.