What Gear Reconcile does
There are three distinct AI actions available on the Gear page:| Action | What it does |
|---|---|
| Find duplicates | Groups equipment items that likely refer to the same piece of gear — different spellings, abbreviations, or naming conventions |
| Suggest canonical value | Reviews conflicting spec values for a gear item and recommends the most accurate one |
| Summarize contributions | Generates a plain-text summary of where team members agree and disagree on a gear item’s specs |
Finding duplicate gear items
When gear is imported from a brief or added at different times, the same piece of equipment can appear in your list under multiple names. For example, “C-Stand”, “C Stand”, and “Century Stand” might all refer to the same item. Gear Reconcile groups these clusters so you can merge them.Run Find duplicates
Click the Reconcile with AI button and select Find duplicates. CineFlow sends your current equipment list to the AI.
Review the duplicate clusters
The AI returns groups of items it believes refer to the same piece of equipment. Each cluster includes a suggested canonical name and a confidence level — high means the items are almost certainly the same, medium means they’re possibly the same and warrant a closer look.
Items imported via Brief Import that were already matched and normalized are excluded from duplicate analysis — only unmatched items are checked.
How AI normalizes gear names
When you run Find duplicates, the AI applies knowledge of common film production equipment naming conventions. It recognizes that abbreviations, brand names, and informal shorthand often refer to the same item:C-Stand/C Stand/Century Stand→ C-StandARRI LF/Alexa LF/Alexa Large Format→ ARRI Alexa LFSennheiser MKH 416/MKH416/416 Shotgun Mic→ Sennheiser MKH 416
Suggesting the correct value for a contested spec
When your gear list has conflicting information for the same item — for example, different team members recorded different sensor sizes or weight figures for the same camera — you can ask the AI to recommend which value is most likely correct.Open the gear item with conflicting data
From the Gear page, open the item that has inconsistent spec data across entries.
Select a field to resolve
Choose the specific field you want the AI to analyze (for example, “sensor size” or “weight”).
Run Suggest canonical value
Click Suggest value for that field. The AI reviews all recorded values and returns a recommendation with a confidence level (low, medium, or high) and a brief explanation.
Summarizing conflicting gear data
For items with many conflicting entries, the Summarize contributions action generates a plain-text overview of where your team agrees and where they differ — without you needing to compare every individual entry manually.Run Summarize contributions
Click Summarize. The AI generates a short summary (under 200 words) covering areas of agreement, fields with conflicting values, and which values appear most frequently.
The summary is labelled as AI-generated wherever it appears in the interface. It reflects only what your team has entered — the AI does not add external information.
Reviewing and accepting AI suggestions
Gear Reconcile is designed around a review step. The AI returns suggestions, and you decide what to apply:- Duplicate clusters: you choose whether to merge, which name to use, and which items to keep
- Spec value suggestions: each suggestion includes a confidence level and a one-sentence explanation of why that value was chosen
- Contribution summaries: read-only — no actions to confirm, just information to guide your review