diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
index ad426234c0..b9fe6e68f1 100644
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@@ -31,6 +31,7 @@ jobs:
           sparse-checkout: |
             .github
             config
+            dashboard
             docs
             instructions
             prompts
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 3ae1e764e3..4c935eee7c 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -23,6 +23,7 @@ checkpoint, and status-only commits are intentionally omitted.
   durable spam audit records without blocking users or mutating repositories.
 - Added a light privacy reminder and stronger screenshot-or-video nudge to real behavior proof review guidance.
 - Added agent-led real behavior proof judgement so ClawSweeper can inspect linked screenshots, videos, logs, and terminal output with a read-only GitHub token, explain the proof verdict in the review comment, tell contributors how to trigger a fresh review after adding proof, and sync `proof: sufficient` when the evidence is convincing.
+- Added a durable review-context budget ledger to generated reports so prompt section sizes, hydrated counts, and truncation state are visible after each run.
 - Added a real behavior proof assessment to PR reviews so missing, mock-only, or insufficient contributor proof blocks pass/automerge markers and asks for screenshots, terminal output, redacted logs, recordings, linked artifacts, or copied live output instead.
 - Added `config/automation-limits.json` plus docs and a drift check so review,
   commit-review, repair, and issue-implementation capacity defaults have one
diff --git a/src/clawsweeper.ts b/src/clawsweeper.ts
index 1b5c812500..4ae8716fe4 100644
--- a/src/clawsweeper.ts
+++ b/src/clawsweeper.ts
@@ -378,6 +378,16 @@ interface ReviewPromptBuild {
   telemetry: ReviewPromptTelemetry;
 }
 
+interface ReviewContextLedgerEntry {
+  section: string;
+  label: string;
+  entries: number;
+  chars: number;
+  total?: number;
+  hydrated?: number;
+  truncated?: boolean;
+}
+
 interface ReviewPromptRuntimeHints {
   proofScratchDir?: string;
 }
@@ -5296,6 +5306,147 @@ function contextCountText(
   return `${displayTotal} (hydrated ${displayHydrated}${truncated ? ", truncated" : ""})`;
 }
 
+function promptJsonChars(value: unknown): number {
+  return JSON.stringify(value, null, 2).length;
+}
+
+function reviewContextLedgerEntry(options: {
+  section: string;
+  label: string;
+  value: unknown;
+  entries: number;
+  total?: number | undefined;
+  hydrated?: number | undefined;
+  truncated?: boolean | undefined;
+}): ReviewContextLedgerEntry {
+  const entry: ReviewContextLedgerEntry = {
+    section: options.section,
+    label: options.label,
+    entries: Math.max(0, Math.round(options.entries)),
+    chars: promptJsonChars(options.value),
+  };
+  if (options.total !== undefined && Number.isFinite(options.total)) {
+    entry.total = Math.max(0, Math.round(options.total));
+  }
+  if (options.hydrated !== undefined && Number.isFinite(options.hydrated)) {
+    entry.hydrated = Math.max(0, Math.round(options.hydrated));
+  }
+  if (options.truncated !== undefined) entry.truncated = options.truncated;
+  return entry;
+}
+
+function arrayEntries(value: unknown[] | undefined): number {
+  return value?.length ?? 0;
+}
+
+function reviewContextLedger(context: ItemContext): ReviewContextLedgerEntry[] {
+  const counts = context.counts;
+  const entries = [
+    reviewContextLedgerEntry({
+      section: "issue",
+      label: "issue",
+      value: context.issue,
+      entries: 1,
+    }),
+    reviewContextLedgerEntry({
+      section: "comments",
+      label: "comments",
+      value: context.comments,
+      entries: context.comments.length,
+      total: counts?.comments,
+      hydrated: counts?.commentsHydrated,
+      truncated: counts?.commentsTruncated,
+    }),
+    reviewContextLedgerEntry({
+      section: "timeline",
+      label: "timeline events",
+      value: context.timeline,
+      entries: context.timeline.length,
+      total: counts?.timeline,
+    }),
+    reviewContextLedgerEntry({
+      section: "closingPullRequests",
+      label: "closing PRs",
+      value: context.closingPullRequests ?? [],
+      entries: arrayEntries(context.closingPullRequests),
+      total: counts?.closingPullRequests,
+    }),
+    reviewContextLedgerEntry({
+      section: "relatedItems",
+      label: "related items",
+      value: context.relatedItems ?? [],
+      entries: arrayEntries(context.relatedItems),
+      total: counts?.relatedItems,
+    }),
+    reviewContextLedgerEntry({
+      section: "pullRequest",
+      label: "pull request",
+      value: context.pullRequest ?? null,
+      entries: context.pullRequest === undefined ? 0 : 1,
+    }),
+    reviewContextLedgerEntry({
+      section: "pullFiles",
+      label: "PR files",
+      value: context.pullFiles ?? [],
+      entries: arrayEntries(context.pullFiles),
+      total: counts?.pullFiles,
+      hydrated: counts?.pullFilesHydrated,
+      truncated: counts?.pullFilesTruncated,
+    }),
+    reviewContextLedgerEntry({
+      section: "pullCommits",
+      label: "PR commits",
+      value: context.pullCommits ?? [],
+      entries: arrayEntries(context.pullCommits),
+      total: counts?.pullCommits,
+      hydrated: counts?.pullCommitsHydrated,
+      truncated: counts?.pullCommitsTruncated,
+    }),
+    reviewContextLedgerEntry({
+      section: "pullReviewComments",
+      label: "PR review comments",
+      value: context.pullReviewComments ?? [],
+      entries: arrayEntries(context.pullReviewComments),
+      total: counts?.pullReviewComments,
+      hydrated: counts?.pullReviewCommentsHydrated,
+      truncated: counts?.pullReviewCommentsTruncated,
+    }),
+    reviewContextLedgerEntry({
+      section: "counts",
+      label: "context counts",
+      value: counts ?? {},
+      entries: Object.keys(counts ?? {}).length,
+    }),
+  ];
+  return entries.filter((entry) => entry.entries > 0 || (entry.total ?? 0) > 0);
+}
+
+export function reviewContextLedgerForTest(context: ItemContext): ReviewContextLedgerEntry[] {
+  return reviewContextLedger(context);
+}
+
+function reviewContextLedgerCountText(entry: ReviewContextLedgerEntry): string {
+  if (entry.total !== undefined || entry.hydrated !== undefined) {
+    const total = entry.total ?? entry.entries;
+    const hydrated = entry.hydrated ?? entry.entries;
+    const suffix = entry.truncated ? ", truncated" : "";
+    return `${hydrated}/${total} hydrated${suffix}`;
+  }
+  return `${entry.entries} ${entry.entries === 1 ? "entry" : "entries"}`;
+}
+
+function renderReviewContextBudget(context: ItemContext): string {
+  return reviewContextLedger(context)
+    .map(
+      (entry) => `- ${entry.label}: ${reviewContextLedgerCountText(entry)}, ${entry.chars} chars`,
+    )
+    .join("\n");
+}
+
+export function renderReviewContextBudgetForTest(context: ItemContext): string {
+  return renderReviewContextBudget(context);
+}
+
 function runtimeReviewTextFromReport(markdown: string): string {
   return runtimeReviewText({
     model: frontMatterValue(markdown, "review_model") ?? "",
@@ -6542,6 +6693,10 @@ ${options.action.closeComment ? options.action.closeComment : "_No close comment
     options.context.counts?.pullReviewCommentsTruncated,
   )}
 
+## Review Context Budget
+
+${renderReviewContextBudget(options.context)}
+
 ## Review Telemetry
 
 - prompt chars: ${reviewTelemetryNumber(options.runtime.promptChars)}
diff --git a/test/clawsweeper.test.ts b/test/clawsweeper.test.ts
index 0d665804bc..6efe9be01b 100644
--- a/test/clawsweeper.test.ts
+++ b/test/clawsweeper.test.ts
@@ -47,7 +47,9 @@ import {
   reviewActionForDecision,
   reviewPriority,
   renderReviewCommentFromReport,
+  renderReviewContextBudgetForTest,
   renderWorkPlanFromReport,
+  reviewContextLedgerForTest,
   reviewDecisionSchemaText,
   reviewPromptTelemetryForTest,
   reviewPromptTemplate,
@@ -465,6 +467,68 @@ test("review prompt telemetry records durable cost proxies", () => {
   assert.equal(telemetry.additionalPromptChars, "keep extra instructions visible".length);
 });
 
+test("review context ledger records ordered section budgets", () => {
+  const context = {
+    issue: { number: 123, title: "Sample PR" },
+    comments: [{ author: "alice", body: "Please review this." }],
+    timeline: [{ event: "committed", sha: "abc123" }],
+    relatedItems: [{ number: 122, title: "Related issue" }],
+    pullRequest: { number: 123, additions: 12 },
+    pullFiles: [
+      { filename: "src/example.ts", patch: "line\n".repeat(20) },
+      { filename: "test/example.test.ts", patch: "test\n".repeat(20) },
+    ],
+    pullCommits: [{ sha: "abc123", message: "fix example" }],
+    pullReviewComments: [],
+    counts: {
+      comments: 10,
+      commentsHydrated: 1,
+      commentsTruncated: true,
+      timeline: 1,
+      relatedItems: 1,
+      pullFiles: 120,
+      pullFilesHydrated: 2,
+      pullFilesTruncated: true,
+      pullCommits: 1,
+      pullCommitsHydrated: 1,
+      pullCommitsTruncated: false,
+      pullReviewComments: 0,
+      pullReviewCommentsHydrated: 0,
+      pullReviewCommentsTruncated: false,
+    },
+  };
+
+  const ledger = reviewContextLedgerForTest(context);
+
+  assert.deepEqual(
+    ledger.map(({ section, entries, total, hydrated, truncated }) => [
+      section,
+      entries,
+      total,
+      hydrated,
+      truncated,
+    ]),
+    [
+      ["issue", 1, undefined, undefined, undefined],
+      ["comments", 1, 10, 1, true],
+      ["timeline", 1, 1, undefined, undefined],
+      ["relatedItems", 1, 1, undefined, undefined],
+      ["pullRequest", 1, undefined, undefined, undefined],
+      ["pullFiles", 2, 120, 2, true],
+      ["pullCommits", 1, 1, 1, false],
+      ["counts", 14, undefined, undefined, undefined],
+    ],
+  );
+  assert.equal(
+    ledger.find((entry) => entry.section === "pullFiles")?.chars,
+    JSON.stringify(context.pullFiles, null, 2).length,
+  );
+  assert.match(
+    renderReviewContextBudgetForTest(context),
+    /- PR files: 2\/120 hydrated, truncated, \d+ chars/,
+  );
+});
+
 test("protected labels are normalized and excluded from normal planning", () => {
   assert.deepEqual(protectedLabels(["Security", "bug", "maintainer", "SECURITY"]), [
     "security",