diff --git a/.env.example b/.env.example
index e0ec46b2..6da63013 100644
--- a/.env.example
+++ b/.env.example
@@ -107,6 +107,8 @@ TZ=UTC
 # Byte budget for images returned by vault_read_asset, in binary bytes before
 # base64 encoding (default: 49152 = 48 KiB, sized for Claude Code's response cap).
 # MAX_IMAGE_OUTPUT_BYTES=49152
+# Max PDF pages to render as images for raw: true (default: 5).
+# MAX_PDF_RENDER_PAGES=5
 # Enable or disable the memory layer (default: true).
 # Set to false to hide memory tools, skip auto-initialization, and omit
 # memory references from server metadata. MEMORY_DIR is ignored when false.
diff --git a/.github/workflows/trivy.yml b/.github/workflows/trivy.yml
index e92920c8..7b7d4730 100644
--- a/.github/workflows/trivy.yml
+++ b/.github/workflows/trivy.yml
@@ -193,7 +193,7 @@ jobs:
         uses: github/codeql-action/upload-sarif@7188fc363630916deb702c7fdcf4e481b751f97a # v4.37.1
         with:
           sarif_file: trivy-results.sarif
-          category: trivy-pr-image
+          category: trivy-published-image
 
       # hashFiles guard: the remote SARIF doesn't exist when the local scan
       # gates the job first (exit-code 1) or the remote build fails.
@@ -202,4 +202,4 @@ jobs:
         uses: github/codeql-action/upload-sarif@7188fc363630916deb702c7fdcf4e481b751f97a # v4.37.1
         with:
           sarif_file: trivy-results-remote.sarif
-          category: trivy-pr-image-remote
+          category: trivy-published-image-remote
diff --git a/AGENTS.md b/AGENTS.md
index b80a8676..b547defb 100644
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -858,7 +858,7 @@ pattern:
 6. **Root compose files** (`docker-compose.yml`, `docker-compose.local.yml`)
    — maintainer/contributor surfaces (if applicable)
 
-CI drift tests in `templates.test.ts` catch omissions across steps 2–4,
+CI drift tests in `cli/src/__tests__/templates.test.ts` catch omissions across steps 2–4,
 but the checklist prevents them.
 
 **Regenerating `social-preview.png`:** Run `npm run render:social-preview`.
diff --git a/ARCHITECTURE.md b/ARCHITECTURE.md
index ead42118..e97eda5b 100644
--- a/ARCHITECTURE.md
+++ b/ARCHITECTURE.md
@@ -296,7 +296,7 @@ Link queries use a `links` table populated during indexing:
 1. **Images** (`.png`/`.jpg`/`.jpeg`/`.gif`/`.webp`) return an MCP `image` content block plus a one-line metadata text block. A shared fit-to-byte-budget pipeline (`utils/fit-image-to-byte-budget.ts`, built on sharp) makes oversized images deliverable: EXIF auto-orient → resize long edge to ≤1568px → walk a fixed quality ladder (JPEG via mozjpeg for opaque images, WebP for alpha — PNG has no quality knob) → shrink dimensions by √(budget/actual) if the ladder floor still exceeds the budget. Deterministic and terminating (bounded attempts, 64px floor); sharp's default `limitInputPixels` stays active as the decompression-bomb guard. The budget (`MAX_IMAGE_OUTPUT_BYTES`, default 48 KiB binary) is sized for the tightest mainstream client cap.
 2. **Canvas** (`.canvas`) linearizes to markdown via the pure `obsidian-markdown/canvas.ts` parser ([JSON Canvas 1.0](https://jsoncanvas.org)): group membership by spatial rect containment (innermost group wins; equal rects tiebreak deterministically by id), nodes in reading order (y, then x), and an edge list with node ids resolved to display names. Lenient parsing — unknown properties ignored, malformed entries skipped. `raw: true` skips the linearizer and returns the JSON source verbatim for full structural fidelity.
 3. **Text formats** (`.svg`/`.json`/`.txt`/`.csv`/`.xml`/`.log`/`.base`) pass through verbatim as text, capped at a fixed 100 KiB output size (explicit error over silent truncation).
-4. **PDFs** (`.pdf`) return structured markdown reconstructed from layout-aware extraction (`unpdf`, based on Mozilla's PDF.js). A document metadata header (title, page count, link count), heading hierarchy inferred from relative font sizes, fenced code blocks detected via monospace fonts, page separators, and a deduplicated links footer. Scanned or image-only PDFs with no extractable text return an error stating the file's existence, size, and page count.
+4. **PDFs** (`.pdf`) return structured markdown reconstructed from layout-aware extraction (`unpdf`, based on Mozilla's PDF.js). A document metadata header (title, page count, link count), heading hierarchy inferred from relative font sizes, fenced code blocks detected via monospace fonts, page separators, and a deduplicated links footer. `raw: true` switches to page-image mode: each page is rendered at 2× scale via `unpdf`'s `renderPageAsImage` with `@napi-rs/canvas` (prebuilt Skia, no system deps), then fitted through the same byte-budget pipeline as regular images. The total image budget is divided evenly across rendered pages (capped at `MAX_PDF_RENDER_PAGES`, default 5). Scanned or image-only PDFs with no extractable text work in raw mode — the model's own vision handles recognition.
 5. **Unknown types** return an error naming the readable set.
 
 The extension-to-representation routing above is implemented by the `vault-operations/asset-operations.ts` use-case. Beneath it, every read goes through `vaultFs.readAsset`, which applies the same `resolveSafePath` traversal guard as notes, rejects `.md` paths (notes belong to `vault_read_note`), and enforces a stat-before-read size cap (`MAX_ASSET_BYTES`, default 50 MiB).
diff --git a/DOCKERHUB.md b/DOCKERHUB.md
index 8979836e..24e6c0e9 100644
--- a/DOCKERHUB.md
+++ b/DOCKERHUB.md
@@ -65,7 +65,7 @@ Your notes embed screenshots, reference architecture diagrams, and link out to c
 
 - **Images** — the image itself, not the filename. Screenshots and diagrams are downscaled and recompressed server-side when they exceed what MCP clients accept, so even a phone session can look at a 5MB architecture diagram
 - **Canvases** — a [Canvas](https://help.obsidian.md/canvas) board arrives as a readable outline: its groups, each card's content in reading order, and the connections between them. The exact JSON source is one flag away when full fidelity matters
-- **PDFs** — text is extracted from the document's content streams; scanned or image-only PDFs return an error stating the page count
+- **PDFs** — text is extracted with heading hierarchy, code blocks, and hyperlinks preserved; set `raw: true` to render pages as images instead, showing layout, diagrams, and tables that text extraction can't preserve — scanned and image-only PDFs work in this mode
 - **Text and data files** — SVG, JSON, CSV, logs, and [Bases](https://help.obsidian.md/bases) files return exactly as written
 - **Browse** — list any folder's assets with per-extension counts and file sizes; assets a note links to report their size in the link graph too
 
@@ -136,25 +136,26 @@ Vault Cortex indexes every [property](https://help.obsidian.md/Editing+and+forma
 
 All settings are environment variables with sensible defaults.
 
-| Variable                    | Required?   | Default                              | Description                                                                                                                                                                                                                       |
-| --------------------------- | ----------- | ------------------------------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
-| `MCP_AUTH_TOKEN`            | Yes         | —                                    | Bearer token for authentication (also the JWT signing key)                                                                                                                                                                        |
-| `VAULT_PATH`                | Local only  | —                                    | Host path to your vault (bind mount source; remote uses a named volume)                                                                                                                                                           |
-| `PUBLIC_URL`                | Remote only | —                                    | Public URL for OAuth discovery metadata                                                                                                                                                                                           |
-| `EMBEDDING_ENABLED`         | —           | `true`                               | Set `false` to disable the embedding pipeline — skips model download, vector tables, embedding passes, and hybrid search. Search falls back to FTS5 keyword matching.                                                             |
-| `RERANK_MODE`               | —           | `blended`                            | Cross-encoder reranking mode: `blended` applies position-aware score blending after RRF fusion (~200ms added latency), `none` skips reranking. Only takes effect when `EMBEDDING_ENABLED` is true.                                |
-| `MEMORY_ENABLED`            | —           | `true`                               | Set `false` to fully disable the memory layer — hides memory tools, skips bootstrap, omits memory from server metadata. `MEMORY_DIR` is ignored when `false`.                                                                     |
-| `MEMORY_DIR`                | —           | `About Me`                           | Vault folder for structured memory files                                                                                                                                                                                          |
-| `PROTECTED_PATHS`           | —           | `MEMORY_DIR, Daily Notes`            | Folders that `vault_delete_note` refuses to touch                                                                                                                                                                                 |
-| `ORPHAN_EXCLUDE_FOLDERS`    | —           | `Daily Notes, Templates, MEMORY_DIR` | Folders excluded from orphan detection                                                                                                                                                                                            |
-| `TZ`                        | —           | `UTC`                                | IANA timezone for timestamps and daily note resolution                                                                                                                                                                            |
-| `SERVICE_DOCUMENTATION_URL` | —           | GitHub repo URL                      | URL returned in OAuth discovery metadata                                                                                                                                                                                          |
-| `LOG_LEVEL`                 | —           | `info`                               | Logging verbosity: `debug`, `info`, `warn`, `error`                                                                                                                                                                               |
-| `LOG_DIR`                   | —           | `/data/logs` (remote), unset (local) | Directory for persistent log files. When set, logs are written to date-stamped files there alongside stdout. Unset means stdout only.                                                                                             |
-| `LOG_RETENTION_DAYS`        | —           | `30`                                 | Days to keep log files before automatic cleanup on startup                                                                                                                                                                        |
-| `WINDOWS_MODE`              | —           | `false`                              | On Windows? Set `true`. Switches the file watcher to polling and note moves to rename-based writes so a vault on a `C:` drive works through Docker Desktop. Safe to leave on for any Windows setup; unneeded on macOS/Linux/WSL2. |
-| `MAX_ASSET_BYTES`           | —           | `52428800` (50 MiB)                  | Maximum file size `vault_read_asset` will read (in bytes). Files exceeding this are rejected before reading. Raise for vaults with very large individual files.                                                                   |
+| Variable                    | Required?   | Default                              | Description                                                                                                                                                                                                                                                     |
+| --------------------------- | ----------- | ------------------------------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| `MCP_AUTH_TOKEN`            | Yes         | —                                    | Bearer token for authentication (also the JWT signing key)                                                                                                                                                                                                      |
+| `VAULT_PATH`                | Local only  | —                                    | Host path to your vault (bind mount source; remote uses a named volume)                                                                                                                                                                                         |
+| `PUBLIC_URL`                | Remote only | —                                    | Public URL for OAuth discovery metadata                                                                                                                                                                                                                         |
+| `EMBEDDING_ENABLED`         | —           | `true`                               | Set `false` to disable the embedding pipeline — skips model download, vector tables, embedding passes, and hybrid search. Search falls back to FTS5 keyword matching.                                                                                           |
+| `RERANK_MODE`               | —           | `blended`                            | Cross-encoder reranking mode: `blended` applies position-aware score blending after RRF fusion (~200ms added latency), `none` skips reranking. Only takes effect when `EMBEDDING_ENABLED` is true.                                                              |
+| `MEMORY_ENABLED`            | —           | `true`                               | Set `false` to fully disable the memory layer — hides memory tools, skips bootstrap, omits memory from server metadata. `MEMORY_DIR` is ignored when `false`.                                                                                                   |
+| `MEMORY_DIR`                | —           | `About Me`                           | Vault folder for structured memory files                                                                                                                                                                                                                        |
+| `PROTECTED_PATHS`           | —           | `MEMORY_DIR, Daily Notes`            | Folders that `vault_delete_note` refuses to touch                                                                                                                                                                                                               |
+| `ORPHAN_EXCLUDE_FOLDERS`    | —           | `Daily Notes, Templates, MEMORY_DIR` | Folders excluded from orphan detection                                                                                                                                                                                                                          |
+| `TZ`                        | —           | `UTC`                                | IANA timezone for timestamps and daily note resolution                                                                                                                                                                                                          |
+| `SERVICE_DOCUMENTATION_URL` | —           | GitHub repo URL                      | URL returned in OAuth discovery metadata                                                                                                                                                                                                                        |
+| `LOG_LEVEL`                 | —           | `info`                               | Logging verbosity: `debug`, `info`, `warn`, `error`                                                                                                                                                                                                             |
+| `LOG_DIR`                   | —           | `/data/logs` (remote), unset (local) | Directory for persistent log files. When set, logs are written to date-stamped files there alongside stdout. Unset means stdout only.                                                                                                                           |
+| `LOG_RETENTION_DAYS`        | —           | `30`                                 | Days to keep log files before automatic cleanup on startup                                                                                                                                                                                                      |
+| `WINDOWS_MODE`              | —           | `false`                              | On Windows? Set `true`. Switches the file watcher to polling and note moves to rename-based writes so a vault on a `C:` drive works through Docker Desktop. Safe to leave on for any Windows setup; unneeded on macOS/Linux/WSL2.                               |
+| `MAX_ASSET_BYTES`           | —           | `52428800` (50 MiB)                  | Maximum file size `vault_read_asset` will read (in bytes). Files exceeding this are rejected before reading. Raise for vaults with very large individual files.                                                                                                 |
 | `MAX_IMAGE_OUTPUT_BYTES`    | —           | `49152` (48 KiB)                     | Byte budget for images delivered by `vault_read_asset`, in binary bytes before base64 encoding. Images exceeding this are downscaled and recompressed to fit. Sized for the tightest mainstream MCP client cap; raise for clients that accept larger responses. |
+| `MAX_PDF_RENDER_PAGES`      | —           | `5`                                  | Maximum PDF pages to render as images when `raw: true` is set on `vault_read_asset`. The per-page byte budget is `MAX_IMAGE_OUTPUT_BYTES` divided evenly across the rendered pages — fewer pages means higher quality each.                                    |
 
 ## Deployment Options
 
diff --git a/README.md b/README.md
index 2c13eb58..083368b8 100644
--- a/README.md
+++ b/README.md
@@ -219,7 +219,7 @@ Your notes embed screenshots, reference architecture diagrams, and link out to c
 
 - **Images** — the image itself, not the filename. Screenshots and diagrams are downscaled and recompressed server-side when they exceed what MCP clients accept, so even a phone session can look at a 5MB architecture diagram
 - **Canvases** — a [Canvas](https://help.obsidian.md/canvas) board arrives as a readable outline: its groups, each card's content in reading order, and the connections between them. The exact JSON source is one flag away when full fidelity matters
-- **PDFs** — text is extracted from the document's content streams; scanned or image-only PDFs return an error stating the page count
+- **PDFs** — text is extracted with heading hierarchy, code blocks, and hyperlinks preserved; set `raw: true` to render pages as images instead, showing layout, diagrams, and tables that text extraction can't preserve — scanned and image-only PDFs work in this mode
 - **Text and data files** — SVG, JSON, CSV, logs, and [Bases](https://help.obsidian.md/bases) files return exactly as written
 - **Browse** — list any folder's assets with per-extension counts and file sizes; assets a note links to report their size in the link graph too
 
@@ -315,6 +315,7 @@ All settings are environment variables with sensible defaults.
 | `WINDOWS_MODE`              | —           | `false`                              | On Windows? Set `true`. Switches the file watcher to polling and note moves to rename-based writes so a vault on a `C:` drive works through Docker Desktop. Safe to leave on for any Windows setup; unneeded on macOS/Linux/WSL2.                               |
 | `MAX_ASSET_BYTES`           | —           | `52428800` (50 MiB)                  | Maximum file size `vault_read_asset` will read (in bytes). Files exceeding this are rejected before reading. Raise for vaults with very large individual files.                                                                                                 |
 | `MAX_IMAGE_OUTPUT_BYTES`    | —           | `49152` (48 KiB)                     | Byte budget for images delivered by `vault_read_asset`, in binary bytes before base64 encoding. Images exceeding this are downscaled and recompressed to fit. Sized for the tightest mainstream MCP client cap; raise for clients that accept larger responses. |
+| `MAX_PDF_RENDER_PAGES`      | —           | `5`                                  | Maximum PDF pages to render as images when `raw: true` is set on `vault_read_asset`. The per-page byte budget is `MAX_IMAGE_OUTPUT_BYTES` divided evenly across the rendered pages — fewer pages means higher quality each.                                     |
 
 **Smart defaults:** Setting `MEMORY_DIR` automatically updates the defaults for `PROTECTED_PATHS` and `ORPHAN_EXCLUDE_FOLDERS`. You only set those explicitly for a fully custom list. When `MEMORY_ENABLED` is `false`, the memory layer is fully disabled — memory tools are hidden and the memory folder is not auto-created.
 
diff --git a/cli/src/env.ts b/cli/src/env.ts
index bf52788f..97c35e88 100644
--- a/cli/src/env.ts
+++ b/cli/src/env.ts
@@ -44,6 +44,11 @@ MAX_ASSET_BYTES=52428800
 # that accept larger tool responses.
 MAX_IMAGE_OUTPUT_BYTES=49152
 
+# Maximum number of PDF pages to render as images when raw: true is set on
+# vault_read_asset (default: 5). The per-page byte budget is MAX_IMAGE_OUTPUT_BYTES
+# divided evenly across the rendered pages. Fewer pages = higher quality each.
+MAX_PDF_RENDER_PAGES=5
+
 # Your IANA timezone — affects daily note resolution and memory timestamps.
 # TZ=America/New_York
 
@@ -148,6 +153,11 @@ MAX_ASSET_BYTES=52428800
 # that accept larger tool responses.
 MAX_IMAGE_OUTPUT_BYTES=49152
 
+# Maximum number of PDF pages to render as images when raw: true is set on
+# vault_read_asset (default: 5). The per-page byte budget is MAX_IMAGE_OUTPUT_BYTES
+# divided evenly across the rendered pages. Fewer pages = higher quality each.
+MAX_PDF_RENDER_PAGES=5
+
 # Enable or disable the memory layer (default: true).
 # Set to false to hide memory tools and skip About Me/ creation.
 MEMORY_ENABLED=true
diff --git a/deploy/local/.env.example b/deploy/local/.env.example
index 500fe76c..ff0e8fb7 100644
--- a/deploy/local/.env.example
+++ b/deploy/local/.env.example
@@ -27,6 +27,11 @@ MAX_ASSET_BYTES=52428800
 # that accept larger tool responses.
 MAX_IMAGE_OUTPUT_BYTES=49152
 
+# Maximum number of PDF pages to render as images when raw: true is set on
+# vault_read_asset (default: 5). The per-page byte budget is MAX_IMAGE_OUTPUT_BYTES
+# divided evenly across the rendered pages. Fewer pages = higher quality each.
+MAX_PDF_RENDER_PAGES=5
+
 # Your IANA timezone — affects daily note resolution and memory timestamps.
 # TZ=America/New_York
 
diff --git a/deploy/local/docker-compose.yml b/deploy/local/docker-compose.yml
index bcfb9d95..6f772138 100644
--- a/deploy/local/docker-compose.yml
+++ b/deploy/local/docker-compose.yml
@@ -38,6 +38,7 @@ services:
       WINDOWS_MODE: ${WINDOWS_MODE:-false}
       MAX_ASSET_BYTES: ${MAX_ASSET_BYTES:-52428800}
       MAX_IMAGE_OUTPUT_BYTES: ${MAX_IMAGE_OUTPUT_BYTES:-49152}
+      MAX_PDF_RENDER_PAGES: ${MAX_PDF_RENDER_PAGES:-5}
       # Optional overrides. When unset, the server applies smart defaults
       # (<MEMORY_DIR> below is the resolved MEMORY_DIR value, default "About Me"):
       #   PROTECTED_PATHS           default: "<MEMORY_DIR>, Daily Notes" (blocked from vault_delete_note)
diff --git a/deploy/remote/.env.example b/deploy/remote/.env.example
index a16c75f5..98bd91ae 100644
--- a/deploy/remote/.env.example
+++ b/deploy/remote/.env.example
@@ -59,6 +59,11 @@ MAX_ASSET_BYTES=52428800
 # that accept larger tool responses.
 MAX_IMAGE_OUTPUT_BYTES=49152
 
+# Maximum number of PDF pages to render as images when raw: true is set on
+# vault_read_asset (default: 5). The per-page byte budget is MAX_IMAGE_OUTPUT_BYTES
+# divided evenly across the rendered pages. Fewer pages = higher quality each.
+MAX_PDF_RENDER_PAGES=5
+
 # Enable or disable the memory layer (default: true).
 # Set to false to hide memory tools and skip About Me/ creation.
 MEMORY_ENABLED=true
diff --git a/deploy/remote/docker-compose.yml b/deploy/remote/docker-compose.yml
index 08a5fadb..951c2470 100644
--- a/deploy/remote/docker-compose.yml
+++ b/deploy/remote/docker-compose.yml
@@ -49,6 +49,7 @@ services:
       WINDOWS_MODE: ${WINDOWS_MODE:-false}
       MAX_ASSET_BYTES: ${MAX_ASSET_BYTES:-52428800}
       MAX_IMAGE_OUTPUT_BYTES: ${MAX_IMAGE_OUTPUT_BYTES:-49152}
+      MAX_PDF_RENDER_PAGES: ${MAX_PDF_RENDER_PAGES:-5}
       # Optional overrides. When unset, the server applies smart defaults
       # (<MEMORY_DIR> below is the resolved MEMORY_DIR value, default "About Me"):
       #   PROTECTED_PATHS           default: "<MEMORY_DIR>, Daily Notes" (blocked from vault_delete_note)
diff --git a/docker-compose.local.yml b/docker-compose.local.yml
index 8821df8a..de757ae9 100644
--- a/docker-compose.local.yml
+++ b/docker-compose.local.yml
@@ -34,6 +34,7 @@ services:
       WINDOWS_MODE: ${WINDOWS_MODE:-false}
       MAX_ASSET_BYTES: ${MAX_ASSET_BYTES:-52428800}
       MAX_IMAGE_OUTPUT_BYTES: ${MAX_IMAGE_OUTPUT_BYTES:-49152}
+      MAX_PDF_RENDER_PAGES: ${MAX_PDF_RENDER_PAGES:-5}
       # Left empty = the server applies smart defaults
       # (<MEMORY_DIR> below is the resolved MEMORY_DIR value, default "About Me"):
       #   PROTECTED_PATHS           default: "<MEMORY_DIR>, Daily Notes"
diff --git a/docker-compose.yml b/docker-compose.yml
index f65e1dc2..7e6166f4 100644
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -47,6 +47,7 @@ services:
       WINDOWS_MODE: ${WINDOWS_MODE:-false}
       MAX_ASSET_BYTES: ${MAX_ASSET_BYTES:-52428800}
       MAX_IMAGE_OUTPUT_BYTES: ${MAX_IMAGE_OUTPUT_BYTES:-49152}
+      MAX_PDF_RENDER_PAGES: ${MAX_PDF_RENDER_PAGES:-5}
       MEMORY_DIR: ${MEMORY_DIR:-About Me}
       # Left empty = the server applies smart defaults
       # (<MEMORY_DIR> below is the resolved MEMORY_DIR value, default "About Me"):
diff --git a/package-lock.json b/package-lock.json
index 9983c90a..837b5217 100644
--- a/package-lock.json
+++ b/package-lock.json
@@ -11,6 +11,7 @@
       "dependencies": {
         "@huggingface/transformers": "4.2.0",
         "@modelcontextprotocol/sdk": "1.29.0",
+        "@napi-rs/canvas": "0.1.100",
         "better-sqlite3": "12.11.1",
         "chokidar": "5.0.0",
         "env-var": "7.5.0",
@@ -2076,6 +2077,270 @@
         }
       }
     },
+    "node_modules/@napi-rs/canvas": {
+      "version": "0.1.100",
+      "resolved": "https://registry.npmjs.org/@napi-rs/canvas/-/canvas-0.1.100.tgz",
+      "integrity": "sha512-xglYA6q3XO5P3BNJYxVZ1IV7DLVjp1Py6nwag88YntrS+3vKHyYcMqXVS4ZztJmwz2uGvz1FWhI/4LgbR5uQDA==",
+      "license": "MIT",
+      "workspaces": [
+        "e2e/*"
+      ],
+      "engines": {
+        "node": ">= 10"
+      },
+      "funding": {
+        "type": "github",
+        "url": "https://github.com/sponsors/Brooooooklyn"
+      },
+      "optionalDependencies": {
+        "@napi-rs/canvas-android-arm64": "0.1.100",
+        "@napi-rs/canvas-darwin-arm64": "0.1.100",
+        "@napi-rs/canvas-darwin-x64": "0.1.100",
+        "@napi-rs/canvas-linux-arm-gnueabihf": "0.1.100",
+        "@napi-rs/canvas-linux-arm64-gnu": "0.1.100",
+        "@napi-rs/canvas-linux-arm64-musl": "0.1.100",
+        "@napi-rs/canvas-linux-riscv64-gnu": "0.1.100",
+        "@napi-rs/canvas-linux-x64-gnu": "0.1.100",
+        "@napi-rs/canvas-linux-x64-musl": "0.1.100",
+        "@napi-rs/canvas-win32-arm64-msvc": "0.1.100",
+        "@napi-rs/canvas-win32-x64-msvc": "0.1.100"
+      }
+    },
+    "node_modules/@napi-rs/canvas-android-arm64": {
+      "version": "0.1.100",
+      "resolved": "https://registry.npmjs.org/@napi-rs/canvas-android-arm64/-/canvas-android-arm64-0.1.100.tgz",
+      "integrity": "sha512-hjhCKhntPv9+t4ckHymdx0phYNcVW+GKQR6Lzw2zE+pOVjOplSmtx9nNNknTjbEDLcuLZqA1y8ufKg1XfgftzQ==",
+      "cpu": [
+        "arm64"
+      ],
+      "license": "MIT",
+      "optional": true,
+      "os": [
+        "android"
+      ],
+      "engines": {
+        "node": ">= 10"
+      },
+      "funding": {
+        "type": "github",
+        "url": "https://github.com/sponsors/Brooooooklyn"
+      }
+    },
+    "node_modules/@napi-rs/canvas-darwin-arm64": {
+      "version": "0.1.100",
+      "resolved": "https://registry.npmjs.org/@napi-rs/canvas-darwin-arm64/-/canvas-darwin-arm64-0.1.100.tgz",
+      "integrity": "sha512-2PcswRaC7Ly645DGt88///zuFDhJxJYdKAs1uU3mfk1atYkXufgcgLfBpk6Tm12nCQBaNt1wpybuPZ4qOhTo8A==",
+      "cpu": [
+        "arm64"
+      ],
+      "license": "MIT",
+      "optional": true,
+      "os": [
+        "darwin"
+      ],
+      "engines": {
+        "node": ">= 10"
+      },
+      "funding": {
+        "type": "github",
+        "url": "https://github.com/sponsors/Brooooooklyn"
+      }
+    },
+    "node_modules/@napi-rs/canvas-darwin-x64": {
+      "version": "0.1.100",
+      "resolved": "https://registry.npmjs.org/@napi-rs/canvas-darwin-x64/-/canvas-darwin-x64-0.1.100.tgz",
+      "integrity": "sha512-ePNZtj7pNIva/siZMg+HmbeozkIjqUIYdoymH8HaA3qK7LfzFN4WMBM8G6HQ9ZC+H3+Dnn5pqtiXpgLykaPOhw==",
+      "cpu": [
+        "x64"
+      ],
+      "license": "MIT",
+      "optional": true,
+      "os": [
+        "darwin"
+      ],
+      "engines": {
+        "node": ">= 10"
+      },
+      "funding": {
+        "type": "github",
+        "url": "https://github.com/sponsors/Brooooooklyn"
+      }
+    },
+    "node_modules/@napi-rs/canvas-linux-arm-gnueabihf": {
+      "version": "0.1.100",
+      "resolved": "https://registry.npmjs.org/@napi-rs/canvas-linux-arm-gnueabihf/-/canvas-linux-arm-gnueabihf-0.1.100.tgz",
+      "integrity": "sha512-d5cDB48oWFGU8/XPhUOFAlySgb/VAu7D+s8fi55K1Pcfg8aPplHWqMgibhVLU8ky7Pyg/fuiVLz4Nf3JrSTuUA==",
+      "cpu": [
+        "arm"
+      ],
+      "license": "MIT",
+      "optional": true,
+      "os": [
+        "linux"
+      ],
+      "engines": {
+        "node": ">= 10"
+      },
+      "funding": {
+        "type": "github",
+        "url": "https://github.com/sponsors/Brooooooklyn"
+      }
+    },
+    "node_modules/@napi-rs/canvas-linux-arm64-gnu": {
+      "version": "0.1.100",
+      "resolved": "https://registry.npmjs.org/@napi-rs/canvas-linux-arm64-gnu/-/canvas-linux-arm64-gnu-0.1.100.tgz",
+      "integrity": "sha512-rDxgxRu69RvDlX/bh9o22DxLsGr8EqsNgotL9+RwQE1S0b0cqeatqsw6aW45mukm0B42DIAaAacKaYQ8cqS1nw==",
+      "cpu": [
+        "arm64"
+      ],
+      "libc": [
+        "glibc"
+      ],
+      "license": "MIT",
+      "optional": true,
+      "os": [
+        "linux"
+      ],
+      "engines": {
+        "node": ">= 10"
+      },
+      "funding": {
+        "type": "github",
+        "url": "https://github.com/sponsors/Brooooooklyn"
+      }
+    },
+    "node_modules/@napi-rs/canvas-linux-arm64-musl": {
+      "version": "0.1.100",
+      "resolved": "https://registry.npmjs.org/@napi-rs/canvas-linux-arm64-musl/-/canvas-linux-arm64-musl-0.1.100.tgz",
+      "integrity": "sha512-K3mDW66N+xT2/V439u1alFANiBUjdEx2gLiNYnCmUsva5jZMxWTjafBYwTzYK+EMFMHrUoabuU+T1BIP5CgbYQ==",
+      "cpu": [
+        "arm64"
+      ],
+      "libc": [
+        "musl"
+      ],
+      "license": "MIT",
+      "optional": true,
+      "os": [
+        "linux"
+      ],
+      "engines": {
+        "node": ">= 10"
+      },
+      "funding": {
+        "type": "github",
+        "url": "https://github.com/sponsors/Brooooooklyn"
+      }
+    },
+    "node_modules/@napi-rs/canvas-linux-riscv64-gnu": {
+      "version": "0.1.100",
+      "resolved": "https://registry.npmjs.org/@napi-rs/canvas-linux-riscv64-gnu/-/canvas-linux-riscv64-gnu-0.1.100.tgz",
+      "integrity": "sha512-mooqUBTIsccZpnoQC4NgrC1v6C1vof39etLNMnBwCY+p0gajWJvAHLGQ6g/gGyS5YrpDW+GefSN4+Cvcr08UWw==",
+      "cpu": [
+        "riscv64"
+      ],
+      "libc": [
+        "glibc"
+      ],
+      "license": "MIT",
+      "optional": true,
+      "os": [
+        "linux"
+      ],
+      "engines": {
+        "node": ">= 10"
+      },
+      "funding": {
+        "type": "github",
+        "url": "https://github.com/sponsors/Brooooooklyn"
+      }
+    },
+    "node_modules/@napi-rs/canvas-linux-x64-gnu": {
+      "version": "0.1.100",
+      "resolved": "https://registry.npmjs.org/@napi-rs/canvas-linux-x64-gnu/-/canvas-linux-x64-gnu-0.1.100.tgz",
+      "integrity": "sha512-1eCvkDCazm7FFhsT7DfGOdSaHgZVK3bt/dSBl5EWHOWmnz+I7j8tPseJqqD81NF+MH21jKUK4wQSDjN0mdhnTg==",
+      "cpu": [
+        "x64"
+      ],
+      "libc": [
+        "glibc"
+      ],
+      "license": "MIT",
+      "optional": true,
+      "os": [
+        "linux"
+      ],
+      "engines": {
+        "node": ">= 10"
+      },
+      "funding": {
+        "type": "github",
+        "url": "https://github.com/sponsors/Brooooooklyn"
+      }
+    },
+    "node_modules/@napi-rs/canvas-linux-x64-musl": {
+      "version": "0.1.100",
+      "resolved": "https://registry.npmjs.org/@napi-rs/canvas-linux-x64-musl/-/canvas-linux-x64-musl-0.1.100.tgz",
+      "integrity": "sha512-20arT6lnI19S68qNlii73TSEDbECNgzMz2EpldC1V3mZFuRkeujXkcebRk0LRJe9SEUAooYiLokfMViY8IX7yA==",
+      "cpu": [
+        "x64"
+      ],
+      "libc": [
+        "musl"
+      ],
+      "license": "MIT",
+      "optional": true,
+      "os": [
+        "linux"
+      ],
+      "engines": {
+        "node": ">= 10"
+      },
+      "funding": {
+        "type": "github",
+        "url": "https://github.com/sponsors/Brooooooklyn"
+      }
+    },
+    "node_modules/@napi-rs/canvas-win32-arm64-msvc": {
+      "version": "0.1.100",
+      "resolved": "https://registry.npmjs.org/@napi-rs/canvas-win32-arm64-msvc/-/canvas-win32-arm64-msvc-0.1.100.tgz",
+      "integrity": "sha512-DZFFT1wIAg37LJw37yhMRFfjATd3vTQzjZ1Yki8u2vhO6Hi5VE6BVaGQ1aaDu7xb4iMErz+9EOwjpS7xcxFeBw==",
+      "cpu": [
+        "arm64"
+      ],
+      "license": "MIT",
+      "optional": true,
+      "os": [
+        "win32"
+      ],
+      "engines": {
+        "node": ">= 10"
+      },
+      "funding": {
+        "type": "github",
+        "url": "https://github.com/sponsors/Brooooooklyn"
+      }
+    },
+    "node_modules/@napi-rs/canvas-win32-x64-msvc": {
+      "version": "0.1.100",
+      "resolved": "https://registry.npmjs.org/@napi-rs/canvas-win32-x64-msvc/-/canvas-win32-x64-msvc-0.1.100.tgz",
+      "integrity": "sha512-MyT1j3mHC2+Lu4pBi9mKyMJhtP6U7k7EldY7sj/uS5gJA65gTXt8MefJQXLJo5d/vZbuWmfxzkEUNc/urV3pHA==",
+      "cpu": [
+        "x64"
+      ],
+      "license": "MIT",
+      "optional": true,
+      "os": [
+        "win32"
+      ],
+      "engines": {
+        "node": ">= 10"
+      },
+      "funding": {
+        "type": "github",
+        "url": "https://github.com/sponsors/Brooooooklyn"
+      }
+    },
     "node_modules/@napi-rs/wasm-runtime": {
       "version": "1.1.6",
       "resolved": "https://registry.npmjs.org/@napi-rs/wasm-runtime/-/wasm-runtime-1.1.6.tgz",
@@ -3844,20 +4109,20 @@
       }
     },
     "node_modules/body-parser": {
-      "version": "2.2.2",
-      "resolved": "https://registry.npmjs.org/body-parser/-/body-parser-2.2.2.tgz",
-      "integrity": "sha512-oP5VkATKlNwcgvxi0vM0p/D3n2C3EReYVX+DNYs5TjZFn/oQt2j+4sVJtSMr18pdRr8wjTcBl6LoV+FUwzPmNA==",
+      "version": "2.3.0",
+      "resolved": "https://registry.npmjs.org/body-parser/-/body-parser-2.3.0.tgz",
+      "integrity": "sha512-2cGmJupaNgg+QUwVLAucDuWuoMZ6EX9iHDRswZ5lsNYEmwPaRknMPCLZz07yTzVq/83p4o/wzbDZbBrTvGGTIw==",
       "license": "MIT",
       "dependencies": {
         "bytes": "^3.1.2",
-        "content-type": "^1.0.5",
+        "content-type": "^2.0.0",
         "debug": "^4.4.3",
-        "http-errors": "^2.0.0",
-        "iconv-lite": "^0.7.0",
+        "http-errors": "^2.0.1",
+        "iconv-lite": "^0.7.2",
         "on-finished": "^2.4.1",
-        "qs": "^6.14.1",
-        "raw-body": "^3.0.1",
-        "type-is": "^2.0.1"
+        "qs": "^6.15.2",
+        "raw-body": "^3.0.2",
+        "type-is": "^2.1.0"
       },
       "engines": {
         "node": ">=18"
@@ -3867,6 +4132,19 @@
         "url": "https://opencollective.com/express"
       }
     },
+    "node_modules/body-parser/node_modules/content-type": {
+      "version": "2.0.0",
+      "resolved": "https://registry.npmjs.org/content-type/-/content-type-2.0.0.tgz",
+      "integrity": "sha512-j/O/d7GcZCyNl7/hwZAb606rzqkyvaDctLmckbxLzHvFBzTJHuGEdodATcP3yIRoDrLHkIATJuvzbFlp/ki2cQ==",
+      "license": "MIT",
+      "engines": {
+        "node": ">=18"
+      },
+      "funding": {
+        "type": "opencollective",
+        "url": "https://opencollective.com/express"
+      }
+    },
     "node_modules/boolean": {
       "version": "3.2.0",
       "resolved": "https://registry.npmjs.org/boolean/-/boolean-3.2.0.tgz",
@@ -7655,17 +7933,34 @@
       }
     },
     "node_modules/type-is": {
-      "version": "2.0.1",
-      "resolved": "https://registry.npmjs.org/type-is/-/type-is-2.0.1.tgz",
-      "integrity": "sha512-OZs6gsjF4vMp32qrCbiVSkrFmXtG/AZhY3t0iAMrMBiAZyV9oALtXO8hsrHbMXF9x6L3grlFuwW2oAz7cav+Gw==",
+      "version": "2.1.0",
+      "resolved": "https://registry.npmjs.org/type-is/-/type-is-2.1.0.tgz",
+      "integrity": "sha512-faYHw0anBbc/kWF3zFTEnxSFOAGUX9GFbOBthvDdLsIlEoWOFOtS0zgCiQYwIskL9iGXZL3kAXD8OoZ4GmMATA==",
       "license": "MIT",
       "dependencies": {
-        "content-type": "^1.0.5",
+        "content-type": "^2.0.0",
         "media-typer": "^1.1.0",
         "mime-types": "^3.0.0"
       },
       "engines": {
-        "node": ">= 0.6"
+        "node": ">= 18"
+      },
+      "funding": {
+        "type": "opencollective",
+        "url": "https://opencollective.com/express"
+      }
+    },
+    "node_modules/type-is/node_modules/content-type": {
+      "version": "2.0.0",
+      "resolved": "https://registry.npmjs.org/content-type/-/content-type-2.0.0.tgz",
+      "integrity": "sha512-j/O/d7GcZCyNl7/hwZAb606rzqkyvaDctLmckbxLzHvFBzTJHuGEdodATcP3yIRoDrLHkIATJuvzbFlp/ki2cQ==",
+      "license": "MIT",
+      "engines": {
+        "node": ">=18"
+      },
+      "funding": {
+        "type": "opencollective",
+        "url": "https://opencollective.com/express"
       }
     },
     "node_modules/typed-query-selector": {
diff --git a/package.json b/package.json
index 28d7848f..d477db4a 100644
--- a/package.json
+++ b/package.json
@@ -73,6 +73,7 @@
     "picomatch": "4.0.5",
     "sharp": "0.35.3",
     "sqlite-vec": "0.1.9",
+    "@napi-rs/canvas": "0.1.100",
     "unpdf": "1.6.2",
     "yaml": "2.9.0",
     "zod": "4.4.3"
diff --git a/server.json b/server.json
index 9e1660cb..8edaf41f 100644
--- a/server.json
+++ b/server.json
@@ -147,6 +147,12 @@
           "description": "Byte budget for images returned by vault_read_asset, in binary bytes before base64 encoding. Images exceeding the budget are downscaled/recompressed server-side to fit; raise for clients that accept larger tool responses.",
           "default": "49152",
           "format": "number"
+        },
+        {
+          "name": "MAX_PDF_RENDER_PAGES",
+          "description": "Maximum PDF pages to render as images when raw: true is set on vault_read_asset. The per-page byte budget is MAX_IMAGE_OUTPUT_BYTES divided evenly across the rendered pages.",
+          "default": "5",
+          "format": "number"
         }
       ]
     }
diff --git a/src/vault-mcp/__tests__/config.test.ts b/src/vault-mcp/__tests__/config.test.ts
index 3a4968fd..b4a69823 100644
--- a/src/vault-mcp/__tests__/config.test.ts
+++ b/src/vault-mcp/__tests__/config.test.ts
@@ -348,4 +348,28 @@ describe("loadConfig", () => {
       )
     })
   })
+
+  describe("MAX_PDF_RENDER_PAGES", () => {
+    it("defaults to 5 when unset", () => {
+      const config = loadConfig(EMPTY_ENV)
+      expect(config.maxPdfRenderPages).toBe(5)
+    })
+
+    it("accepts a custom positive integer", () => {
+      const config = loadConfig({ MAX_PDF_RENDER_PAGES: "10" })
+      expect(config.maxPdfRenderPages).toBe(10)
+    })
+
+    it("rejects a non-integer value", () => {
+      expect(() => loadConfig({ MAX_PDF_RENDER_PAGES: "abc" })).toThrow(
+        /MAX_PDF_RENDER_PAGES/,
+      )
+    })
+
+    it.each(["0", "-1", "1.5"])("rejects non-positive-integer %s", (value) => {
+      expect(() => loadConfig({ MAX_PDF_RENDER_PAGES: value })).toThrow(
+        /MAX_PDF_RENDER_PAGES/,
+      )
+    })
+  })
 })
diff --git a/src/vault-mcp/config.ts b/src/vault-mcp/config.ts
index e0cab0fa..43d1b35c 100644
--- a/src/vault-mcp/config.ts
+++ b/src/vault-mcp/config.ts
@@ -65,6 +65,10 @@ export type VaultConfig = Readonly<{
    *  cap (base64 expands ~4/3, then tokenizes at roughly 3 chars/token).
    *  Set via MAX_IMAGE_OUTPUT_BYTES; raise for clients with looser caps. */
   maxImageOutputBytes: number
+  /** Maximum number of PDF pages to render as images when raw: true is set on
+   *  vault_read_asset. The per-page byte budget is maxImageOutputBytes divided
+   *  evenly across the rendered pages. Set via MAX_PDF_RENDER_PAGES. */
+  maxPdfRenderPages: number
 }>
 
 // ── Loader ─────────────────────────────────────────────────────
@@ -121,7 +125,7 @@ export const loadConfig = (
 
   // env-var's asIntPositive admits 0, but a zero byte cap would make every
   // asset read fail at runtime — reject it at startup instead.
-  const requireNonZeroBytes = (name: string, value: number): number => {
+  const requireNonZero = (name: string, value: number): number => {
     if (value === 0) {
       throw new Error(`env-var: "${name}" must be greater than 0`)
     }
@@ -129,14 +133,14 @@ export const loadConfig = (
   }
 
   // 50 MiB — matches the most permissive prior art for MCP asset reads.
-  const maxAssetBytes = requireNonZeroBytes(
+  const maxAssetBytes = requireNonZero(
     "MAX_ASSET_BYTES",
     envVar.from(env).get("MAX_ASSET_BYTES").default("52428800").asIntPositive(),
   )
 
   // 48 KiB binary ≈ 64 KiB base64 ≈ ~21k tokens — under Claude Code's 25k-token
   // MCP output cap with headroom for the metadata text block.
-  const maxImageOutputBytes = requireNonZeroBytes(
+  const maxImageOutputBytes = requireNonZero(
     "MAX_IMAGE_OUTPUT_BYTES",
     envVar
       .from(env)
@@ -145,6 +149,11 @@ export const loadConfig = (
       .asIntPositive(),
   )
 
+  const maxPdfRenderPages = requireNonZero(
+    "MAX_PDF_RENDER_PAGES",
+    envVar.from(env).get("MAX_PDF_RENDER_PAGES").default("5").asIntPositive(),
+  )
+
   return Object.freeze({
     memoryEnabled,
     memoryDir,
@@ -156,5 +165,6 @@ export const loadConfig = (
     windowsBindMount,
     maxAssetBytes,
     maxImageOutputBytes,
+    maxPdfRenderPages,
   })
 }
diff --git a/src/vault-mcp/mcp-core/tools/asset-tools.ts b/src/vault-mcp/mcp-core/tools/asset-tools.ts
index 5a12760d..af6de087 100644
--- a/src/vault-mcp/mcp-core/tools/asset-tools.ts
+++ b/src/vault-mcp/mcp-core/tools/asset-tools.ts
@@ -2,6 +2,7 @@
 
 import { z } from "zod"
 import { assetOperations } from "../../vault-operations/asset-operations.js"
+import type { AssetReadResult } from "../../vault-operations/asset-operations.js"
 import type { FittedImage } from "../../../utils/fit-image-to-byte-budget.js"
 import type { ToolRegistrationContext } from "./tool-helpers.js"
 import { safeHandler, safeHandlerContent } from "./tool-helpers.js"
@@ -13,6 +14,10 @@ const TOOL_NAMES = {
 
 export { TOOL_NAMES as ASSET_TOOL_NAMES }
 
+type ContentBlock =
+  | { type: "text"; text: string }
+  | { type: "image"; data: string; mimeType: string }
+
 /** One-line, model-facing summary accompanying an image block: what file it
  *  is, what was delivered, and whether/how it was shrunk to fit. */
 const describeDeliveredImage = (result: {
@@ -27,6 +32,45 @@ const describeDeliveredImage = (result: {
   return `${delivered} (recompressed from ${fitted.originalWidth}×${fitted.originalHeight}, ${originalBytes} bytes)`
 }
 
+/** Formats an asset read result into MCP content blocks — the image, pages,
+ *  or text representation the model sees. */
+const formatAssetReadResult = (result: AssetReadResult): ContentBlock[] => {
+  if (result.kind === "image") {
+    return [
+      {
+        type: "image",
+        data: result.fitted.data.toString("base64"),
+        mimeType: result.fitted.mimeType,
+      },
+      { type: "text", text: describeDeliveredImage(result) },
+    ]
+  }
+  if (result.kind === "pages") {
+    const titleSegment = result.title ? `, "${result.title}"` : ""
+    const metadataLine =
+      `${result.path} — PDF, ${result.totalPages} pages` +
+      titleSegment +
+      ` — rendered ${result.pagesRendered} ` +
+      `page${result.pagesRendered === 1 ? "" : "s"} as images`
+    const pageBlocks = result.pages.flatMap((page) => [
+      {
+        type: "image" as const,
+        data: page.fitted.data.toString("base64"),
+        mimeType: page.fitted.mimeType,
+      },
+      {
+        type: "text" as const,
+        text:
+          `Page ${page.pageNumber} — ${page.fitted.mimeType}, ` +
+          `${page.fitted.width}×${page.fitted.height}, ` +
+          `${page.fitted.data.length} bytes`,
+      },
+    ])
+    return [{ type: "text", text: metadataLine }, ...pageBlocks]
+  }
+  return [{ type: "text", text: result.text }]
+}
+
 export const registerAssetTools = ({
   server,
   vaultPath,
@@ -44,11 +88,12 @@ Example: vault_read_asset({ path: "Boards/Roadmap.canvas" }) — a readable outl
 Example: vault_read_asset({ path: "Boards/Roadmap.canvas", raw: true }) — the canvas's exact JSON source
 Example: vault_read_asset({ path: "exports/data.json" }) — the file content as text
 Example: vault_read_asset({ path: "papers/research.pdf" }) — structured text with title, headings, and links
+Example: vault_read_asset({ path: "papers/research.pdf", raw: true }) — each page rendered as an image block
 
 What each type returns:
 - Images (.png/.jpg/.jpeg/.gif/.webp): the image as a viewable image block — downscaled and recompressed server-side when it exceeds client response limits, delivered untouched otherwise — plus a text line stating the path, delivered format/dimensions/bytes, and the original dimensions when shrunk. Animated GIFs are reduced to their first frame when recompressed to fit the budget.
 - Canvas (.canvas): a readable markdown outline per JSON Canvas 1.0 — groups (by visual containment), node content in reading order, and a connections list with edge labels. Set raw: true for the exact JSON source instead (geometry, ids, colors — full fidelity).
-- PDFs (.pdf): structured text with document metadata — title, page count, heading hierarchy (from font sizes), fenced code blocks (from monospace fonts), page separators, and a deduplicated links footer. Richer than flat text extraction: headings, code, and hyperlinks that flat extraction loses are preserved. Scanned or image-only PDFs with no embedded text return an error stating the page count.
+- PDFs (.pdf): structured text with document metadata — title, page count, heading hierarchy (from font sizes), fenced code blocks (from monospace fonts), page separators, and a deduplicated links footer. Richer than flat text extraction: headings, code, and hyperlinks that flat extraction loses are preserved. Set raw: true for page images instead — each page rendered and returned as an image block, showing layout, diagrams, tables, and formatting that text extraction cannot preserve. Image-only and scanned PDFs work in raw mode. Up to ${config.maxPdfRenderPages} pages are rendered.
 - Text formats (.svg/.json/.txt/.csv/.xml/.log/.base): the file content verbatim as text. .svg is returned as its XML source; .base as its YAML source.
 
 When to use: whenever a note references an asset you need to actually see or read — an embedded diagram, a linked canvas, data file, or PDF. Find the assets a note links to (with byte sizes) via vault_get_outgoing_links; browse a folder's assets via vault_list_assets. For .md notes use vault_read_note — this tool rejects them.
@@ -59,12 +104,13 @@ Errors:
 - "asset too large" — the file exceeds the server's read cap (MAX_ASSET_BYTES, default 50 MiB)
 - "text output too large" — a text asset or PDF renders past the output cap; only smaller files can be returned whole
 - "not valid UTF-8" — the file's bytes aren't UTF-8 text; returning them would silently corrupt the content
-- "PDF has no extractable text" — the PDF exists but contains no text content (scanned or image-only); states the page count
+- "PDF has no extractable text" — the PDF exists but contains no text content (scanned or image-only); states the page count. Set raw: true to render pages as images instead
+- "PDF page rendering failed" — raw: true was set but no pages could be rendered; the PDF may be corrupt
 - "image cannot be fitted" — the image could not be compressed under the output budget (MAX_IMAGE_OUTPUT_BYTES)
 - "raw source is not available for images" — raw applies to text-representable files; an image's delivered form is its image block
 - unsupported types (audio, archives, …) return an error naming the readable types plus the file's existence and size
 
-Returns: for images, an image content block plus a one-line metadata text block; for every other supported type, a single text content block.
+Returns: for images, an image content block plus a one-line metadata text block; for PDFs with raw: true, a metadata text block followed by alternating image and text blocks (one pair per page); for every other supported type, a single text content block.
 
 Search coverage: vault_search indexes markdown notes; find assets by browsing (vault_list_assets) or through a note's links (vault_get_outgoing_links).`,
       inputSchema: {
@@ -78,7 +124,7 @@ Search coverage: vault_search indexes markdown notes; find assets by browsing (v
           .boolean()
           .optional()
           .describe(
-            "Return the file's exact text source instead of any rendered form. For .canvas this is the JSON Canvas source (geometry, ids, colors); text formats already return their source, so raw changes nothing there. Images have no text source — raw returns an error.",
+            "Return an alternative representation of the file. For .canvas this is the JSON Canvas source (geometry, ids, colors); for .pdf this renders pages as images instead of extracting text — useful for scanned documents, diagrams, and layout-sensitive content. Text formats already return their source, so raw changes nothing there. Images have no text source — raw returns an error.",
           ),
       },
       annotations: {
@@ -104,6 +150,7 @@ Search coverage: vault_search indexes markdown notes; find assets by browsing (v
               raw,
               maxAssetBytes: config.maxAssetBytes,
               maxImageOutputBytes: config.maxImageOutputBytes,
+              maxPdfRenderPages: config.maxPdfRenderPages,
             },
             reqLogger,
           ),
@@ -120,20 +167,19 @@ Search coverage: vault_search indexes markdown notes; find assets by browsing (v
               originalHeight: result.fitted.originalHeight,
               recompressed: result.fitted.recompressed,
             })
-            return [
-              {
-                type: "image" as const,
-                data: result.fitted.data.toString("base64"),
-                mimeType: result.fitted.mimeType,
-              },
-              { type: "text" as const, text: describeDeliveredImage(result) },
-            ]
+          } else if (result.kind === "pages") {
+            reqLogger.info("tool_result", {
+              path,
+              totalPages: result.totalPages,
+              pagesRendered: result.pagesRendered,
+            })
+          } else {
+            reqLogger.info("tool_result", {
+              path,
+              textBytes: Buffer.byteLength(result.text, "utf8"),
+            })
           }
-          reqLogger.info("tool_result", {
-            path,
-            textBytes: Buffer.byteLength(result.text, "utf8"),
-          })
-          return [{ type: "text" as const, text: result.text }]
+          return formatAssetReadResult(result)
         },
       )
     },
diff --git a/src/vault-mcp/vault-operations/__tests__/asset-operations.test.ts b/src/vault-mcp/vault-operations/__tests__/asset-operations.test.ts
index 790f7b94..e6ce1605 100644
--- a/src/vault-mcp/vault-operations/__tests__/asset-operations.test.ts
+++ b/src/vault-mcp/vault-operations/__tests__/asset-operations.test.ts
@@ -1,4 +1,4 @@
-import { describe, it, expect, vi } from "vitest"
+import { describe, it, expect, vi, beforeEach, onTestFinished } from "vitest"
 import { assetOperations } from "../asset-operations.js"
 import { logger } from "../../../logger.js"
 
@@ -14,14 +14,16 @@ const {
   mockGetMeta,
   mockExtractTextItems,
   mockExtractLinks,
+  mockRenderPageAsImage,
 } = vi.hoisted(() => {
   const mockCleanup = vi.fn()
   return {
     mockCleanup,
-    mockGetDocumentProxy: vi.fn(() => ({ cleanup: mockCleanup })),
+    mockGetDocumentProxy: vi.fn(() => ({ cleanup: mockCleanup, numPages: 1 })),
     mockGetMeta: vi.fn(),
     mockExtractTextItems: vi.fn(),
     mockExtractLinks: vi.fn(),
+    mockRenderPageAsImage: vi.fn(),
   }
 })
 
@@ -30,6 +32,7 @@ vi.mock("unpdf", () => ({
   getMeta: mockGetMeta,
   extractTextItems: mockExtractTextItems,
   extractLinks: mockExtractLinks,
+  renderPageAsImage: mockRenderPageAsImage,
 }))
 
 vi.mock("../../../utils/fit-image-to-byte-budget.js", () => ({
@@ -37,13 +40,16 @@ vi.mock("../../../utils/fit-image-to-byte-budget.js", () => ({
 }))
 
 import { vaultFs } from "../vault-filesystem.js"
+import { fitImageToByteBudget } from "../../../utils/fit-image-to-byte-budget.js"
 
 const mockedReadAsset = vi.mocked(vaultFs.readAsset)
+const mockedFitImage = vi.mocked(fitImageToByteBudget)
 
 const defaultParams = {
   vaultPath: "/vault",
   maxAssetBytes: 52_428_800,
   maxImageOutputBytes: 49_152,
+  maxPdfRenderPages: 5,
 }
 
 /** Builds a single-page StructuredTextItem array from lines of text. Items
@@ -362,6 +368,14 @@ describe("readAssetContent — PDF extraction", () => {
       extension: ".pdf",
     })
     mockGetDocumentProxy.mockRejectedValue(new Error("Invalid PDF structure"))
+    // Restore the default mock regardless of assertion outcome — without
+    // this, a failing assertion leaves subsequent tests with a rejecting mock.
+    onTestFinished(() => {
+      mockGetDocumentProxy.mockResolvedValue({
+        cleanup: mockCleanup,
+        numPages: 1,
+      })
+    })
 
     await expect(
       assetOperations.readAssetContent(
@@ -369,9 +383,6 @@ describe("readAssetContent — PDF extraction", () => {
         logger,
       ),
     ).rejects.toThrow("Invalid PDF structure")
-
-    // Restore the default mock for other tests
-    mockGetDocumentProxy.mockResolvedValue({ cleanup: mockCleanup })
   })
 
   it("cleans up the document proxy after successful extraction", async () => {
@@ -503,3 +514,266 @@ describe("readAssetContent — PDF extraction", () => {
     })
   })
 })
+
+// ── PDF page rendering (raw: true) ────────────────────────────
+
+/** Builds a fake FittedImage result for page rendering tests. */
+const buildFittedImage = (overrides?: {
+  width?: number
+  height?: number
+  dataLength?: number
+}) => ({
+  data: Buffer.alloc(overrides?.dataLength ?? 9_600),
+  mimeType: "image/jpeg",
+  width: overrides?.width ?? 800,
+  height: overrides?.height ?? 1036,
+  originalWidth: 1224,
+  originalHeight: 1584,
+  recompressed: true,
+})
+
+/** Standard PDF mock setup: readAsset returns a .pdf buffer, getMeta
+ *  returns the given title, the proxy reports numPages, and
+ *  extractTextItems is pre-configured (only called in non-raw mode). */
+const setupPdfMocks = (params: { numPages: number; title?: string }) => {
+  mockedReadAsset.mockResolvedValue({
+    buffer: Buffer.from("fake-pdf-bytes"),
+    bytes: 50_000,
+    extension: ".pdf",
+  })
+  mockGetDocumentProxy.mockResolvedValue({
+    cleanup: mockCleanup,
+    numPages: params.numPages,
+  })
+  mockGetMeta.mockResolvedValue({
+    info: params.title ? { Title: params.title } : {},
+  })
+  mockExtractTextItems.mockResolvedValue({
+    totalPages: params.numPages,
+    items: Array.from({ length: params.numPages }, () => []),
+  })
+}
+
+describe("readAssetContent — PDF page rendering (raw: true)", () => {
+  beforeEach(() => {
+    mockRenderPageAsImage.mockReset()
+    mockedFitImage.mockReset()
+    mockCleanup.mockClear()
+    mockGetDocumentProxy.mockReset()
+    mockGetDocumentProxy.mockResolvedValue({
+      cleanup: mockCleanup,
+      numPages: 1,
+    })
+    mockGetMeta.mockReset()
+    mockExtractTextItems.mockReset()
+  })
+
+  it("returns kind pages with rendered images", async () => {
+    setupPdfMocks({ numPages: 2, title: "Visual Doc" })
+    const fakePng = new ArrayBuffer(10_000)
+    mockRenderPageAsImage.mockResolvedValue(fakePng)
+    const fittedResult = buildFittedImage()
+    mockedFitImage.mockResolvedValue(fittedResult)
+
+    const result = await assetOperations.readAssetContent(
+      { ...defaultParams, path: "doc.pdf", raw: true },
+      logger,
+    )
+
+    expect(result).toEqual({
+      kind: "pages",
+      pages: [
+        { pageNumber: 1, fitted: fittedResult, originalBytes: 10_000 },
+        { pageNumber: 2, fitted: fittedResult, originalBytes: 10_000 },
+      ],
+      title: "Visual Doc",
+      totalPages: 2,
+      pagesRendered: 2,
+      path: "doc.pdf",
+    })
+  })
+
+  it("respects maxPdfRenderPages cap", async () => {
+    setupPdfMocks({ numPages: 10, title: "Long PDF" })
+    mockRenderPageAsImage.mockResolvedValue(new ArrayBuffer(5_000))
+    const fittedResult = buildFittedImage()
+    mockedFitImage.mockResolvedValue(fittedResult)
+
+    const result = await assetOperations.readAssetContent(
+      { ...defaultParams, path: "long.pdf", raw: true, maxPdfRenderPages: 3 },
+      logger,
+    )
+
+    expect(result).toMatchObject({
+      kind: "pages",
+      pagesRendered: 3,
+      totalPages: 10,
+      pages: [
+        { pageNumber: 1, fitted: fittedResult },
+        { pageNumber: 2, fitted: fittedResult },
+        { pageNumber: 3, fitted: fittedResult },
+      ],
+    })
+    expect(mockRenderPageAsImage).toHaveBeenCalledTimes(3)
+  })
+
+  it("divides per-page budget evenly across rendered pages", async () => {
+    setupPdfMocks({ numPages: 4 })
+    mockRenderPageAsImage.mockResolvedValue(new ArrayBuffer(1_000))
+    mockedFitImage.mockResolvedValue(buildFittedImage())
+
+    await assetOperations.readAssetContent(
+      {
+        ...defaultParams,
+        path: "budget.pdf",
+        raw: true,
+        maxPdfRenderPages: 4,
+        maxImageOutputBytes: 40_000,
+      },
+      logger,
+    )
+
+    // 40,000 / 4 pages = 10,000 per page
+    for (const call of mockedFitImage.mock.calls) {
+      expect(call[0].budgetBytes).toBe(10_000)
+    }
+  })
+
+  it("skips failed pages and returns the rest", async () => {
+    setupPdfMocks({ numPages: 3, title: "Partial" })
+    mockRenderPageAsImage
+      .mockResolvedValueOnce(new ArrayBuffer(5_000))
+      .mockRejectedValueOnce(new Error("render failed"))
+      .mockResolvedValueOnce(new ArrayBuffer(5_000))
+    const fittedResult = buildFittedImage()
+    mockedFitImage.mockResolvedValue(fittedResult)
+
+    const result = await assetOperations.readAssetContent(
+      {
+        ...defaultParams,
+        path: "partial.pdf",
+        raw: true,
+        maxPdfRenderPages: 3,
+      },
+      logger,
+    )
+
+    expect(result).toEqual({
+      kind: "pages",
+      pages: [
+        { pageNumber: 1, fitted: fittedResult, originalBytes: 5_000 },
+        { pageNumber: 3, fitted: fittedResult, originalBytes: 5_000 },
+      ],
+      title: "Partial",
+      totalPages: 3,
+      pagesRendered: 2,
+      path: "partial.pdf",
+    })
+  })
+
+  it("throws when all pages fail to render", async () => {
+    setupPdfMocks({ numPages: 2 })
+    mockRenderPageAsImage.mockRejectedValue(new Error("render failed"))
+
+    await expect(
+      assetOperations.readAssetContent(
+        {
+          ...defaultParams,
+          path: "broken.pdf",
+          raw: true,
+          maxPdfRenderPages: 2,
+        },
+        logger,
+      ),
+    ).rejects.toThrow(
+      'PDF page rendering failed: "broken.pdf" exists ' +
+        "(50000 bytes, 2 pages) but no pages could be rendered",
+    )
+  })
+
+  it("cleans up the proxy after successful page rendering", async () => {
+    setupPdfMocks({ numPages: 1 })
+    mockRenderPageAsImage.mockResolvedValue(new ArrayBuffer(1_000))
+    mockedFitImage.mockResolvedValue(buildFittedImage())
+
+    await assetOperations.readAssetContent(
+      { ...defaultParams, path: "cleanup.pdf", raw: true },
+      logger,
+    )
+
+    expect(mockCleanup).toHaveBeenCalledOnce()
+  })
+
+  it("cleans up the proxy even when all pages fail", async () => {
+    setupPdfMocks({ numPages: 1 })
+    mockRenderPageAsImage.mockRejectedValue(new Error("render failed"))
+
+    await expect(
+      assetOperations.readAssetContent(
+        { ...defaultParams, path: "fail.pdf", raw: true, maxPdfRenderPages: 1 },
+        logger,
+      ),
+    ).rejects.toThrow("PDF page rendering failed")
+
+    expect(mockCleanup).toHaveBeenCalledOnce()
+  })
+
+  it("passes canvasImport and scale to renderPageAsImage", async () => {
+    setupPdfMocks({ numPages: 1 })
+    mockRenderPageAsImage.mockResolvedValue(new ArrayBuffer(1_000))
+    mockedFitImage.mockResolvedValue(buildFittedImage())
+
+    await assetOperations.readAssetContent(
+      { ...defaultParams, path: "opts.pdf", raw: true },
+      logger,
+    )
+
+    expect(mockRenderPageAsImage).toHaveBeenCalledOnce()
+    expect(mockRenderPageAsImage).toHaveBeenCalledWith(
+      expect.any(Uint8Array),
+      1,
+      expect.objectContaining({
+        canvasImport: expect.any(Function),
+        scale: 2.0,
+      }),
+    )
+  })
+
+  it("does not change text extraction when raw is false", async () => {
+    setupPdfMocks({ numPages: 1, title: "Text Mode" })
+    mockExtractTextItems.mockResolvedValue({
+      totalPages: 1,
+      items: [[...buildPageItems(["Body text"])]],
+    })
+    mockExtractLinks.mockResolvedValue({ links: [], totalPages: 1 })
+
+    const result = await assetOperations.readAssetContent(
+      { ...defaultParams, path: "text.pdf", raw: false },
+      logger,
+    )
+
+    expect(result.kind).toBe("text")
+    expect(mockRenderPageAsImage).not.toHaveBeenCalled()
+  })
+
+  it("omits title from pages result when PDF has no title metadata", async () => {
+    setupPdfMocks({ numPages: 1 })
+    mockRenderPageAsImage.mockResolvedValue(new ArrayBuffer(1_000))
+    const fittedResult = buildFittedImage()
+    mockedFitImage.mockResolvedValue(fittedResult)
+
+    const result = await assetOperations.readAssetContent(
+      { ...defaultParams, path: "notitle.pdf", raw: true },
+      logger,
+    )
+
+    expect(result).toEqual({
+      kind: "pages",
+      pages: [{ pageNumber: 1, fitted: fittedResult, originalBytes: 1_000 }],
+      title: undefined,
+      totalPages: 1,
+      pagesRendered: 1,
+      path: "notitle.pdf",
+    })
+  })
+})
diff --git a/src/vault-mcp/vault-operations/asset-operations.ts b/src/vault-mcp/vault-operations/asset-operations.ts
index ec2f7f3a..b1b109b9 100644
--- a/src/vault-mcp/vault-operations/asset-operations.ts
+++ b/src/vault-mcp/vault-operations/asset-operations.ts
@@ -3,6 +3,7 @@ import {
   getMeta,
   extractTextItems,
   extractLinks,
+  renderPageAsImage,
 } from "unpdf"
 import type { StructuredTextItem } from "unpdf"
 import { vaultFs } from "./vault-filesystem.js"
@@ -53,7 +54,8 @@ const TEXT_PASSTHROUGH_EXTENSIONS = new Set([
 const MAX_TEXT_OUTPUT_BYTES = 102_400
 
 /** The computed result of one asset read, before content-block formatting:
- *  an image (fitted to the byte budget) or a text rendition. */
+ *  an image (fitted to the byte budget), a text rendition, or multiple
+ *  sequential page images (e.g. rendered PDF pages). */
 export type AssetReadResult =
   | Readonly<{
       kind: "image"
@@ -62,6 +64,21 @@ export type AssetReadResult =
       path: string
     }>
   | Readonly<{ kind: "text"; text: string }>
+  /** Multiple sequential page images — e.g. rendered PDF pages. */
+  | Readonly<{
+      kind: "pages"
+      pages: ReadonlyArray<
+        Readonly<{
+          fitted: FittedImage
+          pageNumber: number
+          originalBytes: number
+        }>
+      >
+      title: string | undefined
+      totalPages: number
+      pagesRendered: number
+      path: string
+    }>
 
 /** Rejects text output past the fixed cap — an explicit error beats silent
  *  truncation, and states the actual size so the caller knows what exists. */
@@ -88,6 +105,63 @@ const decodeUtf8Strict = (params: { buffer: Buffer; path: string }): string => {
   }
 }
 
+// ── PDF page rendering ────────────────────────────────────────
+
+/** Render scale for PDF page images — 2.0 produces 1224×1584px for US Letter
+ *  (close to MAX_LONG_EDGE_PX 1568), giving sharp text after JPEG compression
+ *  without wasting pixels that fitImageToByteBudget would discard anyway. */
+const PDF_RENDER_SCALE = 2.0
+
+/** Renders `pagesToRender` pages of a PDF as fitted images, sequentially.
+ *  Individual page failures are logged and skipped — the caller checks whether
+ *  any pages succeeded. Sequential because the unpdf worker can't handle
+ *  concurrent calls on the same proxy (structuredClone error on Node 24).
+ *  Takes raw PDF bytes, not a proxy — renderPageAsImage must create its own
+ *  proxy so pdfjs-dist's internal CanvasFactory is wired to @napi-rs/canvas
+ *  (a proxy created without CanvasFactory falls back to a stub that throws
+ *  on pages needing intermediate canvases for transparency/patterns/masks). */
+const renderPdfPages = async (
+  params: {
+    pdfData: Uint8Array
+    pagesToRender: number
+    perPageBudget: number
+  },
+  logger: Logger,
+): Promise<
+  Array<{ pageNumber: number; fitted: FittedImage; originalBytes: number }>
+> => {
+  const results: Array<{
+    pageNumber: number
+    fitted: FittedImage
+    originalBytes: number
+  }> = []
+
+  for (let pageNumber = 1; pageNumber <= params.pagesToRender; pageNumber++) {
+    try {
+      const pngArrayBuffer = await renderPageAsImage(
+        params.pdfData,
+        pageNumber,
+        {
+          canvasImport: () => import("@napi-rs/canvas"),
+          scale: PDF_RENDER_SCALE,
+        },
+      )
+      const pngBuffer = Buffer.from(pngArrayBuffer)
+      const fitted = await fitImageToByteBudget({
+        buffer: pngBuffer,
+        budgetBytes: params.perPageBudget,
+      })
+      results.push({ pageNumber, fitted, originalBytes: pngBuffer.length })
+    } catch (error) {
+      logger.warn("pdf_page_render_failed", {
+        page: pageNumber,
+        error: String(error),
+      })
+    }
+  }
+  return results
+}
+
 // ── PDF reconstruction ─────────────────────────────────────────
 
 /** Rounds a font size to one decimal place — used as the bucketing key for
@@ -238,6 +312,7 @@ const readAssetContent = async (
     raw?: boolean | undefined
     maxAssetBytes: number
     maxImageOutputBytes: number
+    maxPdfRenderPages: number
   },
   logger: Logger,
 ): Promise<AssetReadResult> => {
@@ -284,6 +359,43 @@ const readAssetContent = async (
       // Sequential — the unpdf worker can't handle concurrent calls
       // on the same proxy (structuredClone error on Node 24).
       const meta = await getMeta(proxy)
+
+      if (raw) {
+        // Page rendering mode — proxy.numPages is a direct getter on
+        // PDFDocumentProxy; avoids extractTextItems which walks every
+        // page to extract text we don't need in raw mode.
+        const totalPages = proxy.numPages
+        const pagesToRender = Math.min(totalPages, params.maxPdfRenderPages)
+        if (pagesToRender === 0) {
+          throw new Error(
+            `PDF page rendering failed: "${path}" exists ` +
+              `(${asset.bytes} bytes) but has 0 pages`,
+          )
+        }
+        const perPageBudget = Math.floor(
+          params.maxImageOutputBytes / pagesToRender,
+        )
+        const pages = await renderPdfPages(
+          { pdfData, pagesToRender, perPageBudget },
+          logger,
+        )
+        if (pages.length === 0) {
+          throw new Error(
+            `PDF page rendering failed: "${path}" exists ` +
+              `(${asset.bytes} bytes, ${totalPages} pages) but no ` +
+              `pages could be rendered`,
+          )
+        }
+        return {
+          kind: "pages",
+          pages,
+          title: meta.info?.Title ?? undefined,
+          totalPages,
+          pagesRendered: pages.length,
+          path,
+        }
+      }
+
       const { totalPages, items } = await extractTextItems(proxy)
       const linkResult = await extractLinks(proxy)