> ## Documentation Index > Fetch the complete documentation index at: https://docs.royaltyport.com/llms.txt > Use this file to discover all available pages before exploring further. # Reviews > Review contracts flagged for low extraction quality and act on repair verdicts ## What are Reviews? The **Reviews** tab surfaces contracts that scored low on extraction quality and need a closer look. When a contract's review score falls below the quality threshold, it is flagged here so you can run the repair agent and decide whether to apply its proposed corrections. ## Accessing Reviews Navigate to **Dashboard** → **Reviews** tab. The tab lists every contract currently flagged for review in the project, with the most actionable items shown first. Hover the **info** icon next to the list heading for a quick reminder of what each status means. ## Overview banner At the top of the Reviews tab, a banner summarizes the health of everything you've ingested into the project across two panels. ### Ingestion The **Ingestion** panel shows how much staged content needs attention, split into two cards: | Card | Covers | | -------------- | --------------------------------------------- | | **Contracts** | Staged contracts moving through the pipeline | | **Statements** | Staged statements moving through the pipeline | Each card shows a **Rejected** count and a **Failed** count — the items that need attention — alongside a progress bar tracking how much of the staged total is not rejected or failed. ### Scores The **Scores** panel summarizes extraction quality across your **Contracts** and **Statements** with a card each. Every card breaks the total into: | Count | Meaning | | ---------- | -------------------------------------------------------------------------------- | | **Good** | Items scoring at or above the quality threshold | | **Review** | Items scoring below the threshold — these are the ones flagged in the list below | Hover the **info** icon next to a score card's title to see the threshold: **Good is 85 or higher, Review is below 85.** ## Review Rows Each flagged contract appears as a collapsible row: | Element | Description | | ------------- | --------------------------------------------------------------------- | | **File name** | The contract name — click it to open the contract in the Drive | | **Score** | The review score at the time of flagging, shown as a percentage badge | | **Status** | Where the contract is in the review lifecycle (see below) | | **Repair** | Runs the repair agent for a flagged contract | Click anywhere on a row (or press **Enter**/**Space**) to expand it and see the triggered rules, the live agent trace, and — once available — the repair verdict. Once a review is **Ready**, its diagnosed **root causes** also appear inline on the row itself, so you can see *why* a contract was flagged without expanding it. A single verdict can list more than one cause — see [Root causes](#root-causes). ### Status | Status | Indicator | Meaning | | ------------- | ----------------- | ------------------------------------------------------------------------------------------------- | | **Flagged** | Flag icon | The contract scored below the quality threshold. Run the repair agent with the **Repair** button. | | **Reviewing** | "Reviewing" badge | The repair agent is running. The row expands automatically so you can watch the live trace. | | **Ready** | Check icon | The agent has finished and produced a verdict awaiting your review. | Only **Flagged** contracts can start a repair run. Once the agent has run for a contract, the row's **Repair** button is disabled. ## Running a repair Locate a row with the **Flagged** status in the Reviews list. Click **Repair** on the row. The status changes to **Reviewing** and the row expands to stream the agent's progress. When the run completes, the contract moves to **Ready** with a verdict you can act on. ## Acting on a verdict When a review is **Ready**, expand the row to see the verdict card. The agent proposes one of two outcomes: | Verdict | What it means | Your action | | ---------- | ------------------------------------------------- | ---------------------------------------------------------------- | | **Repair** | The agent found corrections to the extracted data | Review the proposed changes, then click **Repair** to apply them | | **Ignore** | The agent recommends no change | Click **Ignore** to dismiss the review | The verdict card also shows the agent's reasoning and, for a repair verdict, a summary of the corrected fields so you can confirm the changes before applying them. Both actions require your explicit confirmation — nothing is applied automatically. ### Root causes Alongside the verdict, the card shows one or more **root cause** chips — the agent's diagnosis of *why* the contract was flagged. A single review often spans several causes at once (for example, a missing-context miss and a genuine extraction miss in the same flag), so the agent lists every cause it found rather than forcing one. Chips are ordered with the most significant cause first. The same causes appear inline on the collapsed row, so you can scan them before expanding. Root causes help you spot patterns across your reviews — for example, whether misses come from the extraction itself or from missing context. | Cause | What it means | | -------------------- | ----------------------------------------------------------------------------------------------------------------- | | **Extraction Issue** | The extraction misread evidence it actually had — a genuine extraction miss. | | **Context Issue** | The deciding evidence sat on pages the extraction never saw, so the output was wrong through no fault of its own. | | **Runtime Issue** | The extraction failed mechanically — the output was missing or empty regardless of comprehension. | | **Ambiguity Issue** | The contract itself is ambiguous or silent, so no definitive value exists to extract. | | **Other Issue** | None of the above fit; the agent's reasoning explains what makes the case different. | ## Filters Click **Filter** in the actions bar to narrow the list. A count badge shows how many filter groups are active. | Filter | Options | | ----------- | ---------------------------------------------------------------------------- | | **Status** | Ready, Reviewing, Flagged, Repaired | | **Verdict** | Repair proposed, Ignore proposed | | **Cause** | Extraction Issue, Context Issue, Runtime Issue, Ambiguity Issue, Other Issue | The **Cause** filter matches any-of: a review appears if *any* of its root causes match a selected option, so a multi-cause verdict shows up under each of its causes. Verdict and Cause only exist on **Ready** rows, so selecting either implicitly narrows the list to reviews the agent has already finished. See [Root causes](#root-causes) for what each cause means. Use **Clear** within a group to reset just that group, or **Clear** at the top of the popover to remove all filters at once. ## Sorts Click **Sort** in the actions bar to change the order. Pick a field and a direction: | Sort field | Description | | ---------------- | ---------------------------- | | **Created Date** | When the review was flagged | | **Score** | The review score at flagging | | **Status** | The review lifecycle status | | **File Name** | The contract's file name | Choose **Ascending** or **Descending**, and use **Clear** to return to the default order. By default, reviews are ordered by actionability — items with a verdict awaiting your review come first, then flagged and reviewing items, with the newest first within each group. ## Pagination Long lists are paged. At the bottom of the list you can: * Choose how many reviews to show **per page** (20, 50, or 100) * See the **total** number of flagged contracts * Page through results with the previous, next, and page-number controls Your per-page choice and sort selection are remembered between visits. ## Next Steps Notifications about tasks and changes in your project. Browse and manage all contracts, including their review scores.