diff --git a/docs/swarm/ui.md b/docs/swarm/ui.md
index 34e69c3d..b7ef3651 100644
--- a/docs/swarm/ui.md
+++ b/docs/swarm/ui.md
@@ -16,7 +16,6 @@ domain, covers host-level detail for one hive.
 | `/jobs`                                        | the controller's job graph — where agent creation and credential mints show progress                                    |
 | `/schedules`                                   | the scheduled prompts: add, edit, pause, resume and remove                                                              |
 | `/issues`                                      | a cross-repo issue report                                                                                               |
-| `/review-tryout`                               | try a review classification against a real PR                                                                           |
 
 Everything it shows comes from [`swarm-controller`](../../swarm-controller/README.md).
 An agent created here or with `swarmctl agent create` starts `paused`; set it
@@ -47,27 +46,6 @@ page. Selecting outside it, or Escape, closes it. You can stop or pause the
 agent, or link an account, without leaving the terminal. A subagent's tab opens
 its agent's details.
 
-### Review try-out
-
-`/review-tryout` shows how the [review classifier](review-classifier.md)
-answers a classification on a real PR, before you put it in a repo's config.
-Give it a repo as `owner/name` and a PR number, then write a classification: a
-name, a description, a polarity (positive or negative) and a threshold from 0
-to 100. **add a classification** adds more, and **run** starts the run.
-
-A run takes a minute or more, and up to half an hour for ten classifications.
-The page says it's running and shows how long it has been, then shows the review's table: for each classification the verdict,
-the model's certainty, the threshold, whether the verdict counted and the
-reason. A classification that couldn't run shows its error where the reason
-would be. Under the table are the overall verdict, approve or request changes,
-and the head commit the run read. A failed run shows its error, for example when the swarm
-has no API key yet.
-
-What you typed stays after a run, so you can edit the classification and run it
-again. The page saves no classification and no result, and posts nothing to the
-forge; like every table in the UI, the result table remembers its column
-filters in the browser. It shows no token counts, cache counts or cost.
-
 ### Scheduled prompts
 
 `/schedules` has one card per scheduled prompt. The **name** is the card's
diff --git a/frontend/packages/swarm-ui/src/App.tsx b/frontend/packages/swarm-ui/src/App.tsx
index 44493a1e..bd20222d 100644
--- a/frontend/packages/swarm-ui/src/App.tsx
+++ b/frontend/packages/swarm-ui/src/App.tsx
@@ -8,7 +8,6 @@ import { ComponentsPage } from "./pages/ComponentsPage.js";
 import { JobsPage } from "./pages/JobsPage.js";
 import { HivesPage } from "./pages/HivesPage.js";
 import { IssueReportPage } from "./pages/IssueReportPage.js";
-import { ReviewTryOutPage } from "./pages/review-tryout/ReviewTryOutPage.js";
 import { SchedulesPage } from "./pages/schedules/SchedulesPage.js";
 import { Panel } from "./ui/panel/Panel.js";
 
@@ -34,7 +33,6 @@ export function App() {
         <Route path="/jobs" component={JobsPage} />
         <Route path="/schedules" component={SchedulesPage} />
         <Route path="/issues" component={IssueReportPage} />
-        <Route path="/review-tryout" component={ReviewTryOutPage} />
         <Route path="/components" component={ComponentsPage} />
         <Route component={NotFound} />
       </Switch>
diff --git a/frontend/packages/swarm-ui/src/pages/review-tryout/ClassificationEditor.tsx b/frontend/packages/swarm-ui/src/pages/review-tryout/ClassificationEditor.tsx
deleted file mode 100644
index bdcc2eec..00000000
--- a/frontend/packages/swarm-ui/src/pages/review-tryout/ClassificationEditor.tsx
+++ /dev/null
@@ -1,65 +0,0 @@
-// <ClassificationEditor> — the fields of one review classification: a name, a
-// description (the prompt the model answers), which answer passes, and how
-// certain the model has to be for its answer to count.
-import { Button } from "../../ui/button/Button.js";
-import { SelectField } from "../../ui/select-field/SelectField.js";
-import { TextAreaField } from "../../ui/text-area-field/TextAreaField.js";
-import { TextField } from "../../ui/text-field/TextField.js";
-import type { ClassificationDraft, Polarity } from "./TryOutTypes.js";
-import "./ReviewTryOutPage.css";
-
-const POLARITY_OPTIONS = [
-  { value: "positive", label: "positive — the PR should match" },
-  { value: "negative", label: "negative — the PR should not match" },
-];
-
-export function ClassificationEditor({
-  draft,
-  onChange,
-  onRemove,
-}: {
-  draft: ClassificationDraft;
-  onChange: (next: ClassificationDraft) => void;
-  /** Absent when this is the only classification: one is the minimum. */
-  onRemove?: () => void;
-}) {
-  const id = `tryout-${draft.key}`;
-  return (
-    <div class="ui-tryout-classification">
-      <TextField
-        id={`${id}-name`}
-        label="name"
-        value={draft.name}
-        placeholder="e.g. datetime"
-        onInput={(name) => onChange({ ...draft, name })}
-      />
-      <TextAreaField
-        id={`${id}-description`}
-        label="description"
-        value={draft.description}
-        rows={4}
-        placeholder="what to look for in the PR, e.g. does it add a naive datetime?"
-        onInput={(description) => onChange({ ...draft, description })}
-      />
-      <div class="ui-tryout-classification-row">
-        <SelectField
-          id={`${id}-polarity`}
-          label="polarity"
-          value={draft.polarity}
-          options={POLARITY_OPTIONS}
-          onChange={(polarity) =>
-            onChange({ ...draft, polarity: polarity as Polarity })
-          }
-        />
-        <TextField
-          id={`${id}-threshold`}
-          label="threshold (0–100)"
-          value={draft.threshold}
-          pattern="[0-9]+"
-          onInput={(threshold) => onChange({ ...draft, threshold })}
-        />
-        {onRemove ? <Button onClick={onRemove}>remove</Button> : null}
-      </div>
-    </div>
-  );
-}
diff --git a/frontend/packages/swarm-ui/src/pages/review-tryout/ReviewTryOutPage.css b/frontend/packages/swarm-ui/src/pages/review-tryout/ReviewTryOutPage.css
deleted file mode 100644
index 1d51e603..00000000
--- a/frontend/packages/swarm-ui/src/pages/review-tryout/ReviewTryOutPage.css
+++ /dev/null
@@ -1,66 +0,0 @@
-/* The form panel and the result panel under it, apart from each other. */
-.ui-tryout-page {
-  display: flex;
-  flex-direction: column;
-  gap: 1em;
-}
-/* The try-out form: the repo and PR on one line, then one block per
-   classification, then the actions. */
-.ui-tryout-form {
-  display: flex;
-  flex-direction: column;
-  gap: 1em;
-  margin-top: 1em;
-}
-.ui-tryout-target {
-  display: flex;
-  flex-wrap: wrap;
-  gap: 0.75em;
-}
-.ui-tryout-target > * {
-  flex: 1 1 14em;
-}
-.ui-tryout-classification {
-  display: flex;
-  flex-direction: column;
-  gap: 0.6em;
-  padding: 0.75em;
-  border: 1px solid var(--border);
-  border-radius: 0.5em;
-}
-.ui-tryout-classification-row {
-  display: flex;
-  flex-wrap: wrap;
-  align-items: flex-end;
-  gap: 0.75em;
-}
-.ui-tryout-classification-row > * {
-  flex: 1 1 12em;
-}
-.ui-tryout-classification-row > button {
-  flex: none;
-}
-.ui-tryout-actions {
-  display: flex;
-  flex-wrap: wrap;
-  align-items: center;
-  gap: 0.75em;
-}
-.ui-tryout-hint {
-  margin: 0;
-  color: var(--muted);
-}
-.ui-tryout-running {
-  margin: 0;
-  color: var(--muted);
-}
-.ui-tryout-row-error {
-  color: var(--red);
-}
-.ui-tryout-overall,
-.ui-tryout-sha {
-  margin: 0.75em 0 0;
-}
-.ui-tryout-sha {
-  color: var(--muted);
-}
diff --git a/frontend/packages/swarm-ui/src/pages/review-tryout/ReviewTryOutPage.tsx b/frontend/packages/swarm-ui/src/pages/review-tryout/ReviewTryOutPage.tsx
deleted file mode 100644
index 8c134dfa..00000000
--- a/frontend/packages/swarm-ui/src/pages/review-tryout/ReviewTryOutPage.tsx
+++ /dev/null
@@ -1,239 +0,0 @@
-// <ReviewTryOutPage> — write a review classification and see how the model
-// answers it on a real PR, before it goes into a repo's config
-// (`docs/swarm/review-classifier.md`, "Try-out"). Pick a repo and a PR, write
-// one or more classifications, run: the page posts them to
-// swarm-controller's `POST /api/review-classifier/try-out`, which answers
-// `202 {id}`, and polls `GET .../try-out/{id}` until the run is `done` or
-// `failed`. A run takes a minute or more, so the page says it is running.
-//
-// The inputs stay as typed after a run, so the operator can edit and run
-// again. The page keeps nothing you typed and no result, and posts nothing to
-// the forge; the only thing the browser remembers is the result table's
-// column filters, as for every table in the UI.
-import { useEffect, useState } from "preact/hooks";
-import { ApiErrorPanel } from "@hive/shared/api-error-panel.js";
-import { readApiError, type ProblemDetails } from "@hive/shared/api-error.js";
-import { Button } from "../../ui/button/Button.js";
-import { Panel } from "../../ui/panel/Panel.js";
-import { TextField } from "../../ui/text-field/TextField.js";
-import { ClassificationEditor } from "./ClassificationEditor.js";
-import { TryOutResult } from "./TryOutResult.js";
-import {
-  draftProblem,
-  duplicateName,
-  prNumber,
-  repoProblem,
-  toBody,
-  type ClassificationDraft,
-  type TryOutResult as Result,
-} from "./TryOutTypes.js";
-import "./ReviewTryOutPage.css";
-
-const TRY_OUT = "/api/review-classifier/try-out";
-const POLL_MS = 2000;
-// More classifications than this are not offered a row. It is the try-out
-// endpoint's own cap, which answers a longer list with a 400 that the page
-// shows as it is.
-const MAX_CLASSIFICATIONS = 10;
-
-let nextKey = 0;
-function blankDraft(): ClassificationDraft {
-  return {
-    key: nextKey++,
-    name: "",
-    description: "",
-    polarity: "positive",
-    threshold: "80",
-  };
-}
-
-export function ReviewTryOutPage() {
-  const [repo, setRepo] = useState("");
-  const [pr, setPr] = useState("");
-  const [drafts, setDrafts] = useState<ClassificationDraft[]>(() => [
-    blankDraft(),
-  ]);
-  // The id of the run being polled; null when none is in flight.
-  const [runId, setRunId] = useState<string | null>(null);
-  const [starting, setStarting] = useState(false);
-  const [result, setResult] = useState<Result | null>(null);
-  const [error, setError] = useState<ProblemDetails | null>(null);
-  // Whole seconds since the run started, so a long wait shows it is alive.
-  const [elapsed, setElapsed] = useState(0);
-
-  const running = starting || runId !== null;
-
-  useEffect(() => {
-    if (!running) return;
-    setElapsed(0);
-    const t = setInterval(() => setElapsed((s) => s + 1), 1000);
-    return () => clearInterval(t);
-  }, [running]);
-
-  // Polls the run until it leaves `running`. Cancelled by the next run, or by
-  // leaving the page.
-  useEffect(() => {
-    if (runId === null) return;
-    let cancelled = false;
-    let timer: ReturnType<typeof setTimeout> | undefined;
-    const stop = (problem: ProblemDetails) => {
-      setError(problem);
-      setRunId(null);
-    };
-    async function poll() {
-      try {
-        const res = await fetch(
-          `${TRY_OUT}/${encodeURIComponent(runId ?? "")}`,
-        );
-        if (cancelled) return;
-        if (res.status === 404) {
-          stop({
-            detail:
-              "the result is gone, because it expired or the controller restarted: run it again",
-          });
-          return;
-        }
-        if (!res.ok) {
-          const problem = await readApiError(res);
-          if (!cancelled) stop(problem);
-          return;
-        }
-        const body = (await res.json()) as Result;
-        if (cancelled) return;
-        setResult(body);
-        if (body.state === "running")
-          timer = setTimeout(() => void poll(), POLL_MS);
-        else setRunId(null);
-      } catch (e: unknown) {
-        if (!cancelled) stop({ detail: String(e) });
-      }
-    }
-    void poll();
-    return () => {
-      cancelled = true;
-      if (timer !== undefined) clearTimeout(timer);
-    };
-  }, [runId]);
-
-  const problem =
-    repoProblem(repo) ??
-    (prNumber(pr) === null ? "a PR number is a whole number above 0" : null) ??
-    drafts.map(draftProblem).find((p) => p !== null) ??
-    duplicateName(drafts) ??
-    null;
-
-  async function run(e: Event) {
-    e.preventDefault();
-    const number = prNumber(pr);
-    if (problem !== null || number === null) return;
-    setStarting(true);
-    setError(null);
-    setResult(null);
-    try {
-      const res = await fetch(TRY_OUT, {
-        method: "POST",
-        headers: { "content-type": "application/json" },
-        body: JSON.stringify({
-          repo: repo.trim(),
-          pr: number,
-          classifications: drafts.map(toBody),
-        }),
-      });
-      if (!res.ok) {
-        setError(await readApiError(res));
-        return;
-      }
-      setRunId(((await res.json()) as { id: string }).id);
-    } catch (err: unknown) {
-      setError({ detail: String(err) });
-    } finally {
-      setStarting(false);
-    }
-  }
-
-  function changeDraft(next: ClassificationDraft) {
-    setDrafts((ds) => ds.map((d) => (d.key === next.key ? next : d)));
-  }
-
-  return (
-    <div class="ui-tryout-page">
-      <Panel title="review try-out" icon="🧪">
-        <p>
-          Run a review classification against a real PR and see how the model
-          answers it. Nothing is posted to the forge, and nothing is saved:
-          change the classification and run it again until it answers the way
-          you want.
-        </p>
-        <form class="ui-tryout-form" onSubmit={(e) => void run(e)}>
-          <div class="ui-tryout-target">
-            <TextField
-              id="tryout-repo"
-              label="repo"
-              value={repo}
-              placeholder="owner/name"
-              onInput={setRepo}
-            />
-            <TextField
-              id="tryout-pr"
-              label="PR number"
-              value={pr}
-              pattern="[0-9]+"
-              onInput={setPr}
-            />
-          </div>
-          {drafts.map((d) => (
-            <ClassificationEditor
-              key={d.key}
-              draft={d}
-              onChange={changeDraft}
-              onRemove={
-                drafts.length > 1
-                  ? () => setDrafts((ds) => ds.filter((x) => x.key !== d.key))
-                  : undefined
-              }
-            />
-          ))}
-          <div class="ui-tryout-actions">
-            <Button
-              onClick={() => setDrafts((ds) => [...ds, blankDraft()])}
-              disabled={drafts.length >= MAX_CLASSIFICATIONS}
-            >
-              add a classification
-            </Button>
-            <Button
-              variant="primary"
-              type="submit"
-              disabled={running || problem !== null}
-            >
-              {running ? "running…" : "run"}
-            </Button>
-          </div>
-          {problem !== null ? <p class="ui-tryout-hint">{problem}</p> : null}
-        </form>
-      </Panel>
-      {running || result || error ? (
-        <Panel title="result" icon="📋">
-          {running ? (
-            <p class="ui-tryout-running" role="status">
-              Running — a run takes a minute or more, and up to half an hour for
-              a long list ({elapsed}s so far).
-            </p>
-          ) : null}
-          {error ? (
-            <ApiErrorPanel context="the try-out failed" problem={error} />
-          ) : null}
-          {result?.state === "failed" ? (
-            <ApiErrorPanel
-              context="the run failed"
-              problem={{
-                title: "the classifier could not finish",
-                detail: result.error ?? "no reason given",
-              }}
-            />
-          ) : null}
-          {result?.state === "done" ? <TryOutResult result={result} /> : null}
-        </Panel>
-      ) : null}
-    </div>
-  );
-}
diff --git a/frontend/packages/swarm-ui/src/pages/review-tryout/TryOutResult.tsx b/frontend/packages/swarm-ui/src/pages/review-tryout/TryOutResult.tsx
deleted file mode 100644
index 771f52ec..00000000
--- a/frontend/packages/swarm-ui/src/pages/review-tryout/TryOutResult.tsx
+++ /dev/null
@@ -1,74 +0,0 @@
-// <TryOutResult> — a finished try-out: one row per classification, the
-// overall verdict under them, and the head commit the run read. The columns
-// are the review's own table (`docs/swarm/review-classifier.md`); no token
-// counts, cache counts or cost, here or anywhere in the UI.
-import { Badge } from "@hive/shared/badge.js";
-import { Table, type TableColumn } from "../../ui/table/Table.js";
-import type { TryOutResult as Result, TryOutRow } from "./TryOutTypes.js";
-
-const VERDICT_LABEL: Record<TryOutRow["verdict"], string> = {
-  match: "match",
-  "no-match": "no match",
-  "not-applicable": "not applicable",
-};
-
-const COLUMNS: TableColumn<TryOutRow>[] = [
-  { key: "name", header: "classification", render: (r) => r.name },
-  { key: "polarity", header: "polarity", render: (r) => r.polarity },
-  {
-    key: "verdict",
-    header: "verdict",
-    render: (r) => (r.error ? "—" : VERDICT_LABEL[r.verdict]),
-  },
-  {
-    key: "certainty",
-    header: "certainty",
-    render: (r) => (r.error ? "—" : r.certainty),
-  },
-  { key: "threshold", header: "threshold", render: (r) => r.threshold },
-  {
-    key: "counted",
-    header: "counted",
-    // A counted row shows whether it passed its polarity; one below its
-    // threshold does not count and has no pass or fail.
-    render: (r) =>
-      r.error ? "—" : r.counted ? (r.passed ? "yes ✅" : "yes ❌") : "no",
-  },
-  {
-    key: "reason",
-    header: "reason",
-    cellClass: "ui-table-prose",
-    // A row that could not run says why where the reason would be.
-    render: (r) =>
-      r.error ? <span class="ui-tryout-row-error">{r.error}</span> : r.reason,
-  },
-];
-
-export function TryOutResult({ result }: { result: Result }) {
-  const rows = result.rows ?? [];
-  return (
-    <div class="ui-tryout-result">
-      <Table
-        columns={COLUMNS}
-        rows={rows}
-        // By position: two classifications may share a name, and a shared key
-        // would let one result render over the other.
-        rowKey={(r) => String(rows.indexOf(r))}
-        storageKey="swarm-ui:review-tryout:table-filters"
-        emptyMessage="the run returned no classifications"
-      />
-      {result.overall ? (
-        <p class="ui-tryout-overall">
-          <strong>Overall: </strong>
-          <Badge
-            tone={result.overall === "approve" ? "positive" : "negative"}
-            value={result.overall === "approve" ? "approve" : "request changes"}
-          />
-        </p>
-      ) : null}
-      <p class="ui-tryout-sha">
-        Ran on head <code>{result.head_sha.slice(0, 12)}</code>
-      </p>
-    </div>
-  );
-}
diff --git a/frontend/packages/swarm-ui/src/pages/review-tryout/TryOutTypes.ts b/frontend/packages/swarm-ui/src/pages/review-tryout/TryOutTypes.ts
deleted file mode 100644
index 1114448f..00000000
--- a/frontend/packages/swarm-ui/src/pages/review-tryout/TryOutTypes.ts
+++ /dev/null
@@ -1,108 +0,0 @@
-// Wire shapes of the review classifier's try-out endpoints
-// (`POST /api/review-classifier/try-out`, `GET .../try-out/{id}`), and the
-// checks the form makes before a request is worth sending. The backend stays
-// the real gate: these only keep an obviously bad form from a round trip.
-
-export type Polarity = "positive" | "negative";
-
-/** One classification as the form holds it: every field as the user typed it. */
-export interface ClassificationDraft {
-  /** Local only, so a row keeps its identity when another is removed. */
-  key: number;
-  name: string;
-  description: string;
-  polarity: Polarity;
-  threshold: string;
-}
-
-/** The request's classification: what the backend is sent. */
-export interface ClassificationBody {
-  name: string;
-  description: string;
-  polarity: Polarity;
-  threshold: number;
-}
-
-export type Verdict = "match" | "no-match" | "not-applicable";
-
-export interface TryOutRow {
-  name: string;
-  polarity: Polarity;
-  verdict: Verdict;
-  certainty: number;
-  threshold: number;
-  /** Whether the verdict counted: its certainty reached the threshold. */
-  counted: boolean;
-  /** For a counted row, whether it passed its polarity. */
-  passed: boolean;
-  reason: string;
-  /** Set when this classification could not be run. */
-  error?: string;
-}
-
-export type TryOutOverall = "approve" | "request-changes";
-
-export interface TryOutResult {
-  state: "running" | "done" | "failed";
-  head_sha: string;
-  rows?: TryOutRow[];
-  overall?: TryOutOverall;
-  error?: string;
-}
-
-// Characters the endpoint accepts in each half of a repo: letters, digits, `.`,
-// `_` and `-`, other than `.` and `..` themselves.
-const SEGMENT = /^(?!\.{1,2}$)[A-Za-z0-9._-]+$/;
-
-export function repoProblem(repo: string): string | null {
-  const parts = repo.trim().split("/");
-  return parts.length === 2 && parts.every((p) => SEGMENT.test(p))
-    ? null
-    : "a repo is owner/name, with letters, digits, . _ and - in each part";
-}
-
-export function prNumber(pr: string): number | null {
-  const n = Number(pr.trim());
-  return /^\d+$/.test(pr.trim()) && n > 0 ? n : null;
-}
-
-/** The threshold as a number, or null when it is not a whole 0–100. */
-export function thresholdValue(text: string): number | null {
-  const t = text.trim();
-  if (!/^\d+$/.test(t)) return null;
-  const n = Number(t);
-  return n <= 100 ? n : null;
-}
-
-/** Why a classification is not ready to send, or null when it is. */
-export function draftProblem(d: ClassificationDraft): string | null {
-  if (d.name.trim() === "") return "give each classification a name";
-  if (d.description.trim() === "") {
-    return `${d.name.trim()} needs a description`;
-  }
-  if (thresholdValue(d.threshold) === null) {
-    return `${d.name.trim()}: the threshold is a whole number from 0 to 100`;
-  }
-  return null;
-}
-
-/** A name two classifications share, as a problem, or null: the result table
- * names its rows by it, so a shared name leaves two rows nobody can tell apart. */
-export function duplicateName(drafts: ClassificationDraft[]): string | null {
-  const seen = new Set<string>();
-  for (const d of drafts) {
-    const name = d.name.trim();
-    if (seen.has(name)) return `two classifications are named ${name}`;
-    seen.add(name);
-  }
-  return null;
-}
-
-export function toBody(d: ClassificationDraft): ClassificationBody {
-  return {
-    name: d.name.trim(),
-    description: d.description,
-    polarity: d.polarity,
-    threshold: thresholdValue(d.threshold) ?? 0,
-  };
-}
diff --git a/frontend/packages/swarm-ui/src/shell/Shell.tsx b/frontend/packages/swarm-ui/src/shell/Shell.tsx
index 324d7ad9..7989fc6f 100644
--- a/frontend/packages/swarm-ui/src/shell/Shell.tsx
+++ b/frontend/packages/swarm-ui/src/shell/Shell.tsx
@@ -51,7 +51,6 @@ const NAV_ITEMS: { href: string; label: string; accent: string }[] = [
   { href: "/jobs", label: "jobs", accent: "var(--pink)" },
   { href: "/schedules", label: "schedules", accent: "var(--cyan)" },
   { href: "/issues", label: "issues", accent: "var(--yellow)" },
-  { href: "/review-tryout", label: "review try-out", accent: "var(--red)" },
   { href: "/components", label: "components", accent: "var(--blue)" },
 ];
 
@@ -62,7 +61,7 @@ const NAV_ITEMS: { href: string; label: string; accent: string }[] = [
 // modifier alongside the base class, not a replacement, so every other
 // route's layout is untouched. `/agents` joined for the same reason —
 // its own table view (plus the list+detail split) hits the same cap.
-const WIDE_BODY_ROUTES = new Set(["/issues", "/agents", "/review-tryout"]);
+const WIDE_BODY_ROUTES = new Set(["/issues", "/agents"]);
 
 // `/agents/:name/terminal` (and a subagent's, under
 // `/agents/:name/subagents/:subagent/`) is full-bleed: the terminal fills