diff --git a/frontend/src-tauri/Cargo.lock b/frontend/src-tauri/Cargo.lock
index 17a0db26..6622bab8 100644
--- a/frontend/src-tauri/Cargo.lock
+++ b/frontend/src-tauri/Cargo.lock
@@ -8,15 +8,6 @@ version = "2.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "320119579fcad9c21884f5c4861d16174d0e06250625266f50fe6898340abefa"
 
-[[package]]
-name = "adobe-cmap-parser"
-version = "0.4.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ae8abfa9a4688de8fc9f42b3f013b6fffec18ed8a554f5f113577e0b9b3212a3"
-dependencies = [
- "pom",
-]
-
 [[package]]
 name = "aead"
 version = "0.5.2"
@@ -27,6 +18,17 @@ dependencies = [
  "generic-array",
 ]
 
+[[package]]
+name = "aes"
+version = "0.8.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b169f7a6d4742236a0a00c541b845991d0ac43e546831af1249753ab4c3aa3a0"
+dependencies = [
+ "cfg-if",
+ "cipher",
+ "cpufeatures 0.2.17",
+]
+
 [[package]]
 name = "ahash"
 version = "0.7.8"
@@ -75,7 +77,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "dbb4e440d04be07da1f1bf44fb4495ebd58669372fe0cffa6e48595ac5bd88a3"
 dependencies = [
  "android_log-sys",
- "env_filter",
+ "env_filter 0.1.4",
  "log",
 ]
 
@@ -168,7 +170,7 @@ dependencies = [
  "asn1-rs-derive",
  "asn1-rs-impl",
  "displaydoc",
- "nom",
+ "nom 7.1.3",
  "num-traits",
  "rusticata-macros",
  "thiserror 1.0.69",
@@ -504,6 +506,15 @@ dependencies = [
  "generic-array",
 ]
 
+[[package]]
+name = "block-padding"
+version = "0.3.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a8894febbff9f758034a5b8e12d87918f56dfc64a8e1fe757d65e29041538d93"
+dependencies = [
+ "generic-array",
+]
+
 [[package]]
 name = "block2"
 version = "0.5.1"
@@ -618,6 +629,12 @@ dependencies = [
  "syn 1.0.109",
 ]
 
+[[package]]
+name = "bytecount"
+version = "0.6.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "175812e0be2bccb6abe50bb8d566126198344f707e304f45c648fd8f2cc0365e"
+
 [[package]]
 name = "bytemuck"
 version = "1.24.0"
@@ -630,6 +647,12 @@ version = "1.5.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "1fd0f2584146f6f2ef48085050886acf353beff7305ebd1ae69500e27c67f64b"
 
+[[package]]
+name = "byteorder-lite"
+version = "0.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8f1fe948ff07f4bd06c30984e69f5b4899c516a3ef74f34df92a2df2ab535495"
+
 [[package]]
 name = "bytes"
 version = "1.10.1"
@@ -706,6 +729,15 @@ dependencies = [
  "toml 0.9.8",
 ]
 
+[[package]]
+name = "cbc"
+version = "0.1.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "26b52a9543ae338f279b96b0b9fed9c8093744685043739079ce85cd58f289a6"
+dependencies = [
+ "cipher",
+]
+
 [[package]]
 name = "cc"
 version = "1.2.44"
@@ -763,7 +795,18 @@ checksum = "c3613f74bd2eac03dad61bd53dbe620703d4371614fe0bc3b9f04dd36fe4e818"
 dependencies = [
  "cfg-if",
  "cipher",
- "cpufeatures",
+ "cpufeatures 0.2.17",
+]
+
+[[package]]
+name = "chacha20"
+version = "0.10.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6f8d983286843e49675a4b7a2d174efe136dc93a18d69130dd18198a6c167601"
+dependencies = [
+ "cfg-if",
+ "cpufeatures 0.3.0",
+ "rand_core 0.10.0",
 ]
 
 [[package]]
@@ -773,7 +816,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "10cd79432192d1c0f4e1a0fef9527696cc039165d729fb41b3f4f4f354c2dc35"
 dependencies = [
  "aead",
- "chacha20",
+ "chacha20 0.9.1",
  "cipher",
  "poly1305",
  "zeroize",
@@ -964,6 +1007,15 @@ dependencies = [
  "libc",
 ]
 
+[[package]]
+name = "cpufeatures"
+version = "0.3.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8b2a41393f66f16b0823bb79094d54ac5fbd34ab292ddafb9a0456ac9f87d201"
+dependencies = [
+ "libc",
+]
+
 [[package]]
 name = "crc32fast"
 version = "1.5.0"
@@ -1068,7 +1120,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "97fb8b7c4503de7d6ae7b42ab72a5a59857b4c937ec27a3d4539dba95b5ab2be"
 dependencies = [
  "cfg-if",
- "cpufeatures",
+ "cpufeatures 0.2.17",
  "curve25519-dalek-derive",
  "fiat-crypto",
  "rustc_version",
@@ -1146,7 +1198,7 @@ checksum = "5cd0a5c643689626bec213c4d8bd4d96acc8ffdb4ad4bb6bc16abf27d5f4b553"
 dependencies = [
  "asn1-rs",
  "displaydoc",
- "nom",
+ "nom 7.1.3",
  "num-bigint",
  "num-traits",
  "rusticata-macros",
@@ -1339,6 +1391,15 @@ version = "1.0.20"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "d0881ea181b1df73ff77ffaaf9c7544ecc11e82fba9b5f27b262a3c73a332555"
 
+[[package]]
+name = "ecb"
+version = "0.1.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1a8bfa975b1aec2145850fcaa1c6fe269a16578c44705a532ae3edc92b8881c7"
+dependencies = [
+ "cipher",
+]
+
 [[package]]
 name = "either"
 version = "1.15.0"
@@ -1411,6 +1472,29 @@ dependencies = [
  "regex",
 ]
 
+[[package]]
+name = "env_filter"
+version = "1.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7a1c3cc8e57274ec99de65301228b537f1e4eedc1b8e0f9411c6caac8ae7308f"
+dependencies = [
+ "log",
+ "regex",
+]
+
+[[package]]
+name = "env_logger"
+version = "0.11.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b2daee4ea451f429a58296525ddf28b45a3b64f1acf6587e2067437bb11e218d"
+dependencies = [
+ "anstream",
+ "anstyle",
+ "env_filter 1.0.0",
+ "jiff",
+ "log",
+]
+
 [[package]]
 name = "equivalent"
 version = "1.0.2"
@@ -1438,15 +1522,6 @@ dependencies = [
  "windows-sys 0.61.2",
 ]
 
-[[package]]
-name = "euclid"
-version = "0.20.14"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2bb7ef65b3777a325d1eeefefab5b6d4959da54747e33bd6258e789640f307ad"
-dependencies = [
- "num-traits",
-]
-
 [[package]]
 name = "event-listener"
 version = "5.4.1"
@@ -1475,7 +1550,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "74fef4569247a5f429d9156b9d0a2599914385dd189c539334c625d8099d90ab"
 dependencies = [
  "futures-core",
- "nom",
+ "nom 7.1.3",
  "pin-project-lite",
 ]
 
@@ -1485,6 +1560,26 @@ version = "2.3.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "37909eebbb50d72f9059c3b6d82c0463f2ff062c9e95845c43a6c9c0355411be"
 
+[[package]]
+name = "fax"
+version = "0.2.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f05de7d48f37cd6730705cbca900770cab77a89f413d23e100ad7fad7795a0ab"
+dependencies = [
+ "fax_derive",
+]
+
+[[package]]
+name = "fax_derive"
+version = "0.2.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a0aca10fb742cb43f9e7bb8467c91aa9bcb8e3ffbc6a6f7389bb93ffc920577d"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn 2.0.108",
+]
+
 [[package]]
 name = "fdeflate"
 version = "0.3.7"
@@ -1553,6 +1648,12 @@ version = "1.0.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "3f9eec918d3f24069decb9af1554cad7c880e2da24a9afd88aca000531ab82c1"
 
+[[package]]
+name = "foldhash"
+version = "0.1.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d9c4f5dac5e15c24eb999c26181a6ca40b39fe946cbe4c263c7209467bc83af2"
+
 [[package]]
 name = "foreign-types"
 version = "0.3.2"
@@ -1883,11 +1984,25 @@ dependencies = [
  "cfg-if",
  "js-sys",
  "libc",
- "r-efi",
+ "r-efi 5.3.0",
  "wasip2",
  "wasm-bindgen",
 ]
 
+[[package]]
+name = "getrandom"
+version = "0.4.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0de51e6874e94e7bf76d726fc5d13ba782deca734ff60d5bb2fb2607c7406555"
+dependencies = [
+ "cfg-if",
+ "libc",
+ "r-efi 6.0.0",
+ "rand_core 0.10.0",
+ "wasip2",
+ "wasip3",
+]
+
 [[package]]
 name = "gio"
 version = "0.18.4"
@@ -2061,6 +2176,17 @@ version = "1.8.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "1b43ede17f21864e81be2fa654110bf1e793774238d86ef8555c37e6519c0403"
 
+[[package]]
+name = "half"
+version = "2.7.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6ea2d84b969582b4b1864a92dc5d27cd2b77b622a8d79306834f1be5ba20d84b"
+dependencies = [
+ "cfg-if",
+ "crunchy",
+ "zerocopy",
+]
+
 [[package]]
 name = "hashbrown"
 version = "0.12.3"
@@ -2076,6 +2202,15 @@ version = "0.14.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "e5274423e17b7c9fc20b6e7e208532f9b19825d82dfd615708b70edd83df41f1"
 
+[[package]]
+name = "hashbrown"
+version = "0.15.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9229cfe53dfd69f0609a49f65461bd93001ea1ef889cd5529dd176593f5338a1"
+dependencies = [
+ "foldhash",
+]
+
 [[package]]
 name = "hashbrown"
 version = "0.16.0"
@@ -2293,7 +2428,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "cc50b891e4acf8fe0e71ef88ec43ad82ee07b3810ad09de10f1d01f072ed4b98"
 dependencies = [
  "byteorder",
- "png",
+ "png 0.17.16",
 ]
 
 [[package]]
@@ -2377,6 +2512,12 @@ dependencies = [
  "zerovec",
 ]
 
+[[package]]
+name = "id-arena"
+version = "2.3.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3d3067d79b975e8844ca9eb072e16b31c3c1c36928edf9c6789548c524d0d954"
+
 [[package]]
 name = "ident_case"
 version = "1.0.1"
@@ -2404,6 +2545,22 @@ dependencies = [
  "icu_properties",
 ]
 
+[[package]]
+name = "image"
+version = "0.25.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e6506c6c10786659413faa717ceebcb8f70731c0a60cbae39795fdf114519c1a"
+dependencies = [
+ "bytemuck",
+ "byteorder-lite",
+ "moxcms",
+ "num-traits",
+ "png 0.18.1",
+ "tiff",
+ "zune-core 0.5.1",
+ "zune-jpeg 0.5.12",
+]
+
 [[package]]
 name = "indexmap"
 version = "1.9.3"
@@ -2442,6 +2599,7 @@ version = "0.1.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "879f10e63c20629ecabbb64a8010319738c66a5cd0c29b02d63d272b03751d01"
 dependencies = [
+ "block-padding",
  "generic-array",
 ]
 
@@ -2515,6 +2673,47 @@ dependencies = [
  "system-deps",
 ]
 
+[[package]]
+name = "jiff"
+version = "0.2.22"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "819b44bc7c87d9117eb522f14d46e918add69ff12713c475946b0a29363ed1c2"
+dependencies = [
+ "jiff-static",
+ "jiff-tzdb-platform",
+ "log",
+ "portable-atomic",
+ "portable-atomic-util",
+ "serde_core",
+ "windows-sys 0.61.2",
+]
+
+[[package]]
+name = "jiff-static"
+version = "0.2.22"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "470252db18ecc35fd766c0891b1e3ec6cbbcd62507e85276c01bf75d8e94d4a1"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn 2.0.108",
+]
+
+[[package]]
+name = "jiff-tzdb"
+version = "0.1.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "68971ebff725b9e2ca27a601c5eb38a4c5d64422c4cbab0c535f248087eda5c2"
+
+[[package]]
+name = "jiff-tzdb-platform"
+version = "0.1.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "875a5a69ac2bab1a891711cf5eccbec1ce0341ea805560dcd90b7a2e925132e8"
+dependencies = [
+ "jiff-tzdb",
+]
+
 [[package]]
 name = "jni"
 version = "0.21.1"
@@ -2598,6 +2797,12 @@ version = "1.5.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "bbd2bcb4c963f2ddae06a2efc7e9f3591312473c50c6685e1f298068316e66fe"
 
+[[package]]
+name = "leb128fmt"
+version = "0.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "09edd9e8b54e49e587e4f6295a7d29c3ea94d469cb40ab8ca70b288248a81db2"
+
 [[package]]
 name = "libappindicator"
 version = "0.9.0"
@@ -2687,19 +2892,32 @@ dependencies = [
 
 [[package]]
 name = "lopdf"
-version = "0.34.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c5c8ecfc6c72051981c0459f75ccc585e7ff67c70829560cda8e647882a9abff"
+version = "0.39.0"
+source = "git+https://github.com/J-F-Liu/lopdf?rev=df670a5#df670a5878df541a7a78a909f1555df76d3a12a6"
 dependencies = [
+ "aes",
+ "bitflags 2.10.0",
+ "cbc",
+ "chrono",
+ "ecb",
  "encoding_rs",
  "flate2",
+ "getrandom 0.4.2",
  "indexmap 2.12.0",
  "itoa",
+ "jiff",
  "log",
  "md-5",
- "nom",
+ "nom 8.0.0",
+ "nom_locate",
+ "rand 0.10.0",
  "rangemap",
+ "rayon",
+ "sha2",
+ "stringprep",
+ "thiserror 2.0.17",
  "time",
+ "ttf-parser",
  "weezl",
 ]
 
@@ -2723,15 +2941,18 @@ dependencies = [
  "axum",
  "base64 0.22.1",
  "dirs 5.0.1",
+ "flate2",
  "futures-util",
  "hound",
+ "image",
  "log",
+ "lopdf",
  "maple-proxy",
  "ndarray",
  "once_cell",
  "openssl",
  "ort",
- "pdf-extract",
+ "pdf-inspector",
  "plist",
  "rand 0.8.5",
  "rand_distr",
@@ -2898,6 +3119,16 @@ dependencies = [
  "windows-sys 0.61.2",
 ]
 
+[[package]]
+name = "moxcms"
+version = "0.7.11"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ac9557c559cd6fc9867e122e20d2cbefc9ca29d80d027a8e39310920ed2f0a97"
+dependencies = [
+ "num-traits",
+ "pxfm",
+]
+
 [[package]]
 name = "muda"
 version = "0.17.1"
@@ -2913,7 +3144,7 @@ dependencies = [
  "objc2-core-foundation",
  "objc2-foundation 0.3.2",
  "once_cell",
- "png",
+ "png 0.17.16",
  "serde",
  "thiserror 2.0.17",
  "windows-sys 0.60.2",
@@ -3017,6 +3248,26 @@ dependencies = [
  "minimal-lexical",
 ]
 
+[[package]]
+name = "nom"
+version = "8.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "df9761775871bdef83bee530e60050f7e54b1105350d6884eb0fb4f46c2f9405"
+dependencies = [
+ "memchr",
+]
+
+[[package]]
+name = "nom_locate"
+version = "5.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0b577e2d69827c4740cba2b52efaad1c4cc7c73042860b199710b3575c68438d"
+dependencies = [
+ "bytecount",
+ "memchr",
+ "nom 8.0.0",
+]
+
 [[package]]
 name = "nu-ansi-term"
 version = "0.50.3"
@@ -3642,18 +3893,18 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "df94ce210e5bc13cb6651479fa48d14f601d9858cfe0467f43ae157023b938d3"
 
 [[package]]
-name = "pdf-extract"
-version = "0.7.12"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "cbb3a5387b94b9053c1e69d8abfd4dd6dae7afda65a5c5279bc1f42ab39df575"
+name = "pdf-inspector"
+version = "0.1.0"
+source = "git+https://github.com/firecrawl/pdf-inspector?branch=main#2b5fcd7bbb33dd9840281b97f1ca6244e8660cdf"
 dependencies = [
- "adobe-cmap-parser",
- "encoding_rs",
- "euclid",
+ "env_logger",
+ "log",
  "lopdf",
- "postscript",
- "type1-encoding-parser",
- "unicode-normalization",
+ "once_cell",
+ "rayon",
+ "regex",
+ "thiserror 2.0.17",
+ "ttf-parser",
 ]
 
 [[package]]
@@ -3880,6 +4131,19 @@ dependencies = [
  "miniz_oxide",
 ]
 
+[[package]]
+name = "png"
+version = "0.18.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "60769b8b31b2a9f263dae2776c37b1b28ae246943cf719eb6946a1db05128a61"
+dependencies = [
+ "bitflags 2.10.0",
+ "crc32fast",
+ "fdeflate",
+ "flate2",
+ "miniz_oxide",
+]
+
 [[package]]
 name = "polling"
 version = "3.11.0"
@@ -3900,17 +4164,11 @@ version = "0.8.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "8159bd90725d2df49889a078b54f4f79e87f1f8a8444194cdca81d38f5393abf"
 dependencies = [
- "cpufeatures",
+ "cpufeatures 0.2.17",
  "opaque-debug",
  "universal-hash",
 ]
 
-[[package]]
-name = "pom"
-version = "1.1.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "60f6ce597ecdcc9a098e7fddacb1065093a3d66446fa16c675e7e71d1b5c28e6"
-
 [[package]]
 name = "portable-atomic"
 version = "1.11.1"
@@ -3926,12 +4184,6 @@ dependencies = [
  "portable-atomic",
 ]
 
-[[package]]
-name = "postscript"
-version = "0.14.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "78451badbdaebaf17f053fd9152b3ffb33b516104eacb45e7864aaa9c712f306"
-
 [[package]]
 name = "potential_utf"
 version = "0.1.4"
@@ -3962,6 +4214,16 @@ version = "0.1.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "925383efa346730478fb4838dbe9137d2a47675ad789c546d150a6e1dd4ab31c"
 
+[[package]]
+name = "prettyplease"
+version = "0.2.37"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "479ca8adacdd7ce8f1fb39ce9ecccbfe93a3f1344b3d0d97f20bc0196208f62b"
+dependencies = [
+ "proc-macro2",
+ "syn 2.0.108",
+]
+
 [[package]]
 name = "proc-macro-crate"
 version = "1.3.1"
@@ -4050,6 +4312,18 @@ dependencies = [
  "syn 1.0.109",
 ]
 
+[[package]]
+name = "pxfm"
+version = "0.1.28"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b5a041e753da8b807c9255f28de81879c78c876392ff2469cde94799b2896b9d"
+
+[[package]]
+name = "quick-error"
+version = "2.0.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a993555f31e5a609f617c12db6250dedcac1b0a85076912c436e6fc9b2c8e6a3"
+
 [[package]]
 name = "quick-xml"
 version = "0.38.3"
@@ -4129,6 +4403,12 @@ version = "5.3.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "69cdb34c158ceb288df11e18b4bd39de994f6657d83847bdffdbd7f346754b0f"
 
+[[package]]
+name = "r-efi"
+version = "6.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f8dcc9c7d52a811697d2151c701e0d08956f92b0e24136cf4cf27b57a6a0d9bf"
+
 [[package]]
 name = "radium"
 version = "0.7.0"
@@ -4170,6 +4450,17 @@ dependencies = [
  "rand_core 0.9.3",
 ]
 
+[[package]]
+name = "rand"
+version = "0.10.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "bc266eb313df6c5c09c1c7b1fbe2510961e5bcd3add930c1e31f7ed9da0feff8"
+dependencies = [
+ "chacha20 0.10.0",
+ "getrandom 0.4.2",
+ "rand_core 0.10.0",
+]
+
 [[package]]
 name = "rand_chacha"
 version = "0.2.2"
@@ -4227,6 +4518,12 @@ dependencies = [
  "getrandom 0.3.4",
 ]
 
+[[package]]
+name = "rand_core"
+version = "0.10.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0c8d0fd677905edcbeedbf2edb6494d676f0e98d54d5cf9bda0b061cb8fb8aba"
+
 [[package]]
 name = "rand_distr"
 version = "0.4.3"
@@ -4519,7 +4816,7 @@ version = "4.1.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "faf0c4a6ece9950b9abdb62b1cfcf2a68b3b67a10ba445b3bb85be2a293d0632"
 dependencies = [
- "nom",
+ "nom 7.1.3",
 ]
 
 [[package]]
@@ -4742,7 +5039,7 @@ version = "0.11.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "2bef2ebfde456fb76bbcf9f59315333decc4fda0b2b44b420243c11e0f5ec1f5"
 dependencies = [
- "half",
+ "half 1.8.3",
  "serde",
 ]
 
@@ -4912,7 +5209,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "a7507d819769d01a365ab707794a4084392c824f54a7a6a7862f8c3d0892b283"
 dependencies = [
  "cfg-if",
- "cpufeatures",
+ "cpufeatures 0.2.17",
  "digest",
 ]
 
@@ -5088,6 +5385,17 @@ dependencies = [
  "quote",
 ]
 
+[[package]]
+name = "stringprep"
+version = "0.1.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7b4df3d392d81bd458a8a621b8bffbd2302a12ffe288a9d931670948749463b1"
+dependencies = [
+ "unicode-bidi",
+ "unicode-normalization",
+ "unicode-properties",
+]
+
 [[package]]
 name = "strsim"
 version = "0.11.1"
@@ -5354,7 +5662,7 @@ dependencies = [
  "ico",
  "json-patch",
  "plist",
- "png",
+ "png 0.17.16",
  "proc-macro2",
  "quote",
  "semver",
@@ -5739,6 +6047,20 @@ dependencies = [
  "cfg-if",
 ]
 
+[[package]]
+name = "tiff"
+version = "0.10.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "af9605de7fee8d9551863fd692cce7637f548dbd9db9180fcc07ccc6d26c336f"
+dependencies = [
+ "fax",
+ "flate2",
+ "half 2.7.1",
+ "quick-error",
+ "weezl",
+ "zune-jpeg 0.4.21",
+]
+
 [[package]]
 name = "time"
 version = "0.3.44"
@@ -6092,7 +6414,7 @@ dependencies = [
  "objc2-core-graphics",
  "objc2-foundation 0.3.2",
  "once_cell",
- "png",
+ "png 0.17.16",
  "serde",
  "thiserror 2.0.17",
  "windows-sys 0.60.2",
@@ -6105,13 +6427,10 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "e421abadd41a4225275504ea4d6566923418b7f05506fbc9c0fe86ba7396114b"
 
 [[package]]
-name = "type1-encoding-parser"
-version = "0.1.0"
+name = "ttf-parser"
+version = "0.25.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d3d6cc09e1a99c7e01f2afe4953789311a1c50baebbdac5b477ecf78e2e92a5b"
-dependencies = [
- "pom",
-]
+checksum = "d2df906b07856748fa3f6e0ad0cbaa047052d4a7dd609e231c4f72cee8c36f31"
 
 [[package]]
 name = "typeid"
@@ -6177,6 +6496,12 @@ dependencies = [
  "unic-common",
 ]
 
+[[package]]
+name = "unicode-bidi"
+version = "0.3.18"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5c1cb5db39152898a79168971543b1cb5020dff7fe43c8dc468b0885f5e29df5"
+
 [[package]]
 name = "unicode-ident"
 version = "1.0.22"
@@ -6192,12 +6517,24 @@ dependencies = [
  "tinyvec",
 ]
 
+[[package]]
+name = "unicode-properties"
+version = "0.1.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7df058c713841ad818f1dc5d3fd88063241cc61f49f5fbea4b951e8cf5a8d71d"
+
 [[package]]
 name = "unicode-segmentation"
 version = "1.12.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "f6ccf251212114b54433ec949fd6a7841275f9ada20dddd2f29e9ceea4501493"
 
+[[package]]
+name = "unicode-xid"
+version = "0.2.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ebc1c04c71510c7f702b52b7c350734c9ff1295c464a03335b00bb84fc54f853"
+
 [[package]]
 name = "universal-hash"
 version = "0.5.1"
@@ -6391,7 +6728,16 @@ version = "1.0.1+wasi-0.2.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "0562428422c63773dad2c345a1882263bbf4d65cf3f42e90921f787ef5ad58e7"
 dependencies = [
- "wit-bindgen",
+ "wit-bindgen 0.46.0",
+]
+
+[[package]]
+name = "wasip3"
+version = "0.4.0+wasi-0.3.0-rc-2026-01-06"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5428f8bf88ea5ddc08faddef2ac4a67e390b88186c703ce6dbd955e1c145aca5"
+dependencies = [
+ "wit-bindgen 0.51.0",
 ]
 
 [[package]]
@@ -6452,6 +6798,28 @@ dependencies = [
  "unicode-ident",
 ]
 
+[[package]]
+name = "wasm-encoder"
+version = "0.244.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "990065f2fe63003fe337b932cfb5e3b80e0b4d0f5ff650e6985b1048f62c8319"
+dependencies = [
+ "leb128fmt",
+ "wasmparser",
+]
+
+[[package]]
+name = "wasm-metadata"
+version = "0.244.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "bb0e353e6a2fbdc176932bbaab493762eb1255a7900fe0fea1a2f96c296cc909"
+dependencies = [
+ "anyhow",
+ "indexmap 2.12.0",
+ "wasm-encoder",
+ "wasmparser",
+]
+
 [[package]]
 name = "wasm-streams"
 version = "0.4.2"
@@ -6465,6 +6833,18 @@ dependencies = [
  "web-sys",
 ]
 
+[[package]]
+name = "wasmparser"
+version = "0.244.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "47b807c72e1bac69382b3a6fb3dbe8ea4c0ed87ff5629b8685ae6b9a611028fe"
+dependencies = [
+ "bitflags 2.10.0",
+ "hashbrown 0.15.5",
+ "indexmap 2.12.0",
+ "semver",
+]
+
 [[package]]
 name = "web-sys"
 version = "0.3.82"
@@ -7134,6 +7514,94 @@ version = "0.46.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "f17a85883d4e6d00e8a97c586de764dabcc06133f7f1d55dce5cdc070ad7fe59"
 
+[[package]]
+name = "wit-bindgen"
+version = "0.51.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d7249219f66ced02969388cf2bb044a09756a083d0fab1e566056b04d9fbcaa5"
+dependencies = [
+ "wit-bindgen-rust-macro",
+]
+
+[[package]]
+name = "wit-bindgen-core"
+version = "0.51.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ea61de684c3ea68cb082b7a88508a8b27fcc8b797d738bfc99a82facf1d752dc"
+dependencies = [
+ "anyhow",
+ "heck 0.5.0",
+ "wit-parser",
+]
+
+[[package]]
+name = "wit-bindgen-rust"
+version = "0.51.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b7c566e0f4b284dd6561c786d9cb0142da491f46a9fbed79ea69cdad5db17f21"
+dependencies = [
+ "anyhow",
+ "heck 0.5.0",
+ "indexmap 2.12.0",
+ "prettyplease",
+ "syn 2.0.108",
+ "wasm-metadata",
+ "wit-bindgen-core",
+ "wit-component",
+]
+
+[[package]]
+name = "wit-bindgen-rust-macro"
+version = "0.51.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0c0f9bfd77e6a48eccf51359e3ae77140a7f50b1e2ebfe62422d8afdaffab17a"
+dependencies = [
+ "anyhow",
+ "prettyplease",
+ "proc-macro2",
+ "quote",
+ "syn 2.0.108",
+ "wit-bindgen-core",
+ "wit-bindgen-rust",
+]
+
+[[package]]
+name = "wit-component"
+version = "0.244.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9d66ea20e9553b30172b5e831994e35fbde2d165325bec84fc43dbf6f4eb9cb2"
+dependencies = [
+ "anyhow",
+ "bitflags 2.10.0",
+ "indexmap 2.12.0",
+ "log",
+ "serde",
+ "serde_derive",
+ "serde_json",
+ "wasm-encoder",
+ "wasm-metadata",
+ "wasmparser",
+ "wit-parser",
+]
+
+[[package]]
+name = "wit-parser"
+version = "0.244.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ecc8ac4bc1dc3381b7f59c34f00b67e18f910c2c0f50015669dde7def656a736"
+dependencies = [
+ "anyhow",
+ "id-arena",
+ "indexmap 2.12.0",
+ "log",
+ "semver",
+ "serde",
+ "serde_derive",
+ "serde_json",
+ "unicode-xid",
+ "wasmparser",
+]
+
 [[package]]
 name = "writeable"
 version = "0.6.2"
@@ -7237,7 +7705,7 @@ dependencies = [
  "data-encoding",
  "der-parser",
  "lazy_static",
- "nom",
+ "nom 7.1.3",
  "oid-registry",
  "rusticata-macros",
  "thiserror 1.0.69",
@@ -7450,6 +7918,36 @@ dependencies = [
  "memchr",
 ]
 
+[[package]]
+name = "zune-core"
+version = "0.4.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3f423a2c17029964870cfaabb1f13dfab7d092a62a29a89264f4d36990ca414a"
+
+[[package]]
+name = "zune-core"
+version = "0.5.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "cb8a0807f7c01457d0379ba880ba6322660448ddebc890ce29bb64da71fb40f9"
+
+[[package]]
+name = "zune-jpeg"
+version = "0.4.21"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "29ce2c8a9384ad323cf564b67da86e21d3cfdff87908bc1223ed5c99bc792713"
+dependencies = [
+ "zune-core 0.4.12",
+]
+
+[[package]]
+name = "zune-jpeg"
+version = "0.5.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "410e9ecef634c709e3831c2cfdb8d9c32164fae1c67496d5b68fff728eec37fe"
+dependencies = [
+ "zune-core 0.5.1",
+]
+
 [[package]]
 name = "zvariant"
 version = "5.8.0"
diff --git a/frontend/src-tauri/Cargo.toml b/frontend/src-tauri/Cargo.toml
index 5b166c97..6a1714c8 100644
--- a/frontend/src-tauri/Cargo.toml
+++ b/frontend/src-tauri/Cargo.toml
@@ -37,7 +37,10 @@ maple-proxy = "0.1.7"
 tauri-plugin-fs = "2.4.4"
 anyhow = "1.0"
 axum = "0.8"
-pdf-extract = "0.7"
+pdf-inspector = { git = "https://github.com/firecrawl/pdf-inspector", branch = "main" }
+lopdf = { git = "https://github.com/J-F-Liu/lopdf", rev = "df670a5" }
+image = { version = "0.25", default-features = false, features = ["tiff", "jpeg", "png"] }
+flate2 = "1"
 base64 = "0.22"
 
 [target.'cfg(any(target_os = "macos", target_os = "linux", target_os = "windows"))'.dependencies]
diff --git a/frontend/src-tauri/src/pdf_extractor.rs b/frontend/src-tauri/src/pdf_extractor.rs
index 2b652a24..fa9ec0ab 100644
--- a/frontend/src-tauri/src/pdf_extractor.rs
+++ b/frontend/src-tauri/src/pdf_extractor.rs
@@ -1,11 +1,16 @@
 use base64::{engine::general_purpose::STANDARD as BASE64, Engine};
-use pdf_extract::extract_text_from_mem;
+use pdf_inspector::{process_pdf_mem, PdfType};
 use serde::{Deserialize, Serialize};
 
 #[derive(Debug, Serialize, Deserialize)]
 pub struct DocumentData {
     pub filename: String,
     pub text_content: String,
+    /// Base64-encoded page images (data URLs) for scanned/image-based PDFs.
+    /// Present when the PDF needs OCR -- the frontend should send these to a
+    /// vision model (e.g. Qwen 3 VL) for text extraction.
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub page_images: Option<Vec<String>>,
 }
 
 #[derive(Debug, Serialize, Deserialize)]
@@ -25,33 +30,377 @@ pub async fn extract_document_content(
         .decode(&file_base64)
         .map_err(|e| format!("Failed to decode base64 file: {e}"))?;
 
-    let text_content = match file_type.as_str() {
+    match file_type.as_str() {
         "pdf" | "application/pdf" => {
-            // Extract text from PDF
-            extract_text_from_mem(&file_bytes)
-                .map_err(|e| format!("Failed to extract text from PDF: {e}"))?
+            // Use pdf-inspector for full PDF processing (detect + extract + markdown)
+            let result = process_pdf_mem(&file_bytes)
+                .map_err(|e| format!("Failed to extract text from PDF: {e}"))?;
+
+            // Log PDF classification info for debugging
+            log::info!(
+                "PDF '{filename}': type={:?}, pages={}, confidence={:.2}, has_encoding_issues={}, time={}ms",
+                result.pdf_type,
+                result.page_count,
+                result.confidence,
+                result.has_encoding_issues,
+                result.processing_time_ms
+            );
+
+            if result.has_encoding_issues {
+                log::warn!(
+                    "PDF '{filename}' has encoding issues - extracted text may be incomplete or garbled"
+                );
+            }
+
+            // pdf-inspector extracts what it can from all PDF types.
+            // For scanned/image-based PDFs it may return limited or no text.
+            let markdown = result.markdown.unwrap_or_default();
+
+            if markdown.trim().is_empty() {
+                // For scanned/image-based PDFs, extract page images for vision-model OCR
+                if matches!(result.pdf_type, PdfType::Scanned | PdfType::ImageBased) {
+                    log::info!(
+                        "PDF '{filename}' is scanned/image-based -- extracting page images for OCR"
+                    );
+                    let page_images = extract_page_images(&file_bytes)?;
+                    if page_images.is_empty() {
+                        return Err(
+                            "This PDF appears to be scanned or image-based but no page images could be extracted."
+                                .to_string(),
+                        );
+                    }
+                    return Ok(DocumentResponse {
+                        document: DocumentData {
+                            filename,
+                            text_content: String::new(),
+                            page_images: Some(page_images),
+                        },
+                        status: "completed".to_string(),
+                    });
+                }
+                return Err("No text content could be extracted from this PDF.".to_string());
+            }
+
+            Ok(DocumentResponse {
+                document: DocumentData {
+                    filename,
+                    text_content: markdown,
+                    page_images: None,
+                },
+                status: "completed".to_string(),
+            })
         }
         "txt" | "text/plain" | "md" | "text/markdown" => {
             // For text files, just convert bytes to string
-            String::from_utf8(file_bytes).map_err(|e| format!("Failed to decode text file: {e}"))?
+            let text_content = String::from_utf8(file_bytes)
+                .map_err(|e| format!("Failed to decode text file: {e}"))?;
+            Ok(DocumentResponse {
+                document: DocumentData {
+                    filename,
+                    text_content,
+                    page_images: None,
+                },
+                status: "completed".to_string(),
+            })
+        }
+        _ => Err(format!("Unsupported file type: {file_type}")),
+    }
+}
+
+// ---------------------------------------------------------------------------
+// Scanned-PDF page image extraction
+// ---------------------------------------------------------------------------
+
+/// Extract embedded images from each page of a scanned PDF.
+///
+/// Returns a Vec of base64 data-URL strings (e.g. `data:image/jpeg;base64,...`).
+/// Handles the most common image encodings found in scanned documents:
+///   - DCTDecode (JPEG) -- raw bytes passed through
+///   - CCITTFaxDecode (Group 3/4) -- wrapped in a minimal TIFF, decoded, re-encoded as JPEG
+///   - FlateDecode (raw pixels) -- decompressed and encoded as JPEG
+fn extract_page_images(pdf_bytes: &[u8]) -> Result<Vec<String>, String> {
+    let doc = lopdf::Document::load_mem(pdf_bytes)
+        .map_err(|e| format!("Failed to parse PDF for image extraction: {e}"))?;
+
+    let mut images: Vec<String> = Vec::new();
+
+    for (_page_num, page_id) in doc.get_pages() {
+        let page = doc
+            .get_object(page_id)
+            .map_err(|e| format!("Failed to get page object: {e}"))?;
+        let dict = match page.as_dict() {
+            Ok(d) => d,
+            Err(_) => continue,
+        };
+
+        let resources = match dict.get(b"Resources") {
+            Ok(r) => match doc.dereference(r) {
+                Ok((_, obj)) => obj.clone(),
+                Err(_) => continue,
+            },
+            Err(_) => continue,
+        };
+
+        let xobjects = match resources.as_dict().and_then(|d| d.get(b"XObject")) {
+            Ok(x) => match doc.dereference(x) {
+                Ok((_, obj)) => obj.clone(),
+                Err(_) => continue,
+            },
+            Err(_) => continue,
+        };
+
+        let xobj_dict = match xobjects.as_dict() {
+            Ok(d) => d,
+            Err(_) => continue,
+        };
+
+        for (_name, obj_ref) in xobj_dict.iter() {
+            let obj = match doc.dereference(obj_ref) {
+                Ok((_, o)) => o.clone(),
+                Err(_) => continue,
+            };
+
+            let stream = match obj.as_stream() {
+                Ok(s) => s,
+                Err(_) => continue,
+            };
+
+            // Only process Image XObjects
+            let subtype = stream
+                .dict
+                .get(b"Subtype")
+                .and_then(|s| s.as_name())
+                .unwrap_or(b"");
+            if subtype != b"Image" {
+                continue;
+            }
+
+            // Determine the filter (compression) type.
+            // Per PDF spec §7.3.4.2, Filter can be a single Name or an Array of Names.
+            // In an array, the last filter describes the final data format (e.g.
+            // [/ASCII85Decode /DCTDecode] means the underlying data is JPEG).
+            let filter_name: Vec<u8> = stream
+                .dict
+                .get(b"Filter")
+                .ok()
+                .and_then(|f| {
+                    f.as_name().map(|n| n.to_vec()).ok().or_else(|| {
+                        f.as_array()
+                            .ok()
+                            .and_then(|arr| arr.last())
+                            .and_then(|last| last.as_name().ok())
+                            .map(|n| n.to_vec())
+                    })
+                })
+                .unwrap_or_default();
+
+            let data_url = match filter_name.as_slice() {
+                b"DCTDecode" => {
+                    // Raw JPEG -- pass through directly
+                    let b64 = BASE64.encode(&stream.content);
+                    format!("data:image/jpeg;base64,{b64}")
+                }
+                b"CCITTFaxDecode" => {
+                    // CCITT Group 3/4 fax data -- wrap in TIFF, decode, re-encode as JPEG
+                    decode_ccitt_to_data_url(stream)?
+                }
+                b"FlateDecode" => {
+                    // Raw pixel data compressed with zlib
+                    decode_flate_to_data_url(stream)?
+                }
+                other => {
+                    let name = String::from_utf8_lossy(other);
+                    log::warn!("Unsupported PDF image filter '{name}' -- skipping");
+                    continue;
+                }
+            };
+
+            images.push(data_url);
+        }
+    }
+
+    Ok(images)
+}
+
+/// Decode a CCITTFaxDecode image stream by wrapping it in a minimal TIFF file,
+/// then decoding with the `image` crate and re-encoding as JPEG.
+fn decode_ccitt_to_data_url(stream: &lopdf::Stream) -> Result<String, String> {
+    let width =
+        get_stream_int(&stream.dict, b"Width").ok_or("CCITTFaxDecode image missing Width")?;
+    let height =
+        get_stream_int(&stream.dict, b"Height").ok_or("CCITTFaxDecode image missing Height")?;
+
+    // K < 0 = Group 4, K = 0 = Group 3 1-D, K > 0 = Group 3 2-D
+    let k = get_decode_parm_int(&stream.dict, b"K").unwrap_or(-1);
+
+    let tiff_bytes = wrap_ccitt_as_tiff(&stream.content, width as u32, height as u32, k as i32);
+
+    let img = image::load_from_memory(&tiff_bytes)
+        .map_err(|e| format!("Failed to decode CCITT image via TIFF wrapper: {e}"))?;
+
+    encode_image_to_jpeg_data_url(&img)
+}
+
+/// Decode a FlateDecode (zlib-compressed raw pixels) image stream.
+fn decode_flate_to_data_url(stream: &lopdf::Stream) -> Result<String, String> {
+    let width =
+        get_stream_int(&stream.dict, b"Width").ok_or("FlateDecode image missing Width")? as u32;
+    let height =
+        get_stream_int(&stream.dict, b"Height").ok_or("FlateDecode image missing Height")? as u32;
+    let bpc = get_stream_int(&stream.dict, b"BitsPerComponent").unwrap_or(8) as u32;
+
+    // Decompress zlib data
+    use std::io::Read;
+    let mut decoder = flate2::read::ZlibDecoder::new(&stream.content[..]);
+    let mut raw_pixels = Vec::new();
+    decoder
+        .read_to_end(&mut raw_pixels)
+        .map_err(|e| format!("Failed to decompress FlateDecode image: {e}"))?;
+
+    // Determine color space
+    let cs_name = stream
+        .dict
+        .get(b"ColorSpace")
+        .and_then(|c| c.as_name().map(|n| n.to_vec()))
+        .unwrap_or_else(|_| b"DeviceGray".to_vec());
+
+    let img: image::DynamicImage = match (cs_name.as_slice(), bpc) {
+        (b"DeviceGray", 8) => {
+            let gray = image::GrayImage::from_raw(width, height, raw_pixels)
+                .ok_or("Failed to construct grayscale image from raw pixels")?;
+            image::DynamicImage::ImageLuma8(gray)
+        }
+        (b"DeviceRGB", 8) => {
+            let rgb = image::RgbImage::from_raw(width, height, raw_pixels)
+                .ok_or("Failed to construct RGB image from raw pixels")?;
+            image::DynamicImage::ImageRgb8(rgb)
+        }
+        (b"DeviceGray", 1) => {
+            // 1-bit grayscale -- expand to 8-bit
+            let expanded = expand_1bit_to_8bit(&raw_pixels, width, height);
+            let gray = image::GrayImage::from_raw(width, height, expanded)
+                .ok_or("Failed to construct 1-bit grayscale image")?;
+            image::DynamicImage::ImageLuma8(gray)
         }
         _ => {
-            return Err(format!("Unsupported file type: {file_type}"));
+            let name = String::from_utf8_lossy(&cs_name);
+            return Err(format!(
+                "Unsupported FlateDecode color space / bpc: {name}/{bpc}"
+            ));
         }
     };
 
-    Ok(DocumentResponse {
-        document: DocumentData {
-            filename,
-            text_content,
-        },
-        status: "completed".to_string(),
-    })
+    encode_image_to_jpeg_data_url(&img)
+}
+
+/// Encode a DynamicImage as a JPEG data URL.
+fn encode_image_to_jpeg_data_url(img: &image::DynamicImage) -> Result<String, String> {
+    let mut jpeg_buf = Vec::new();
+    let mut cursor = std::io::Cursor::new(&mut jpeg_buf);
+    img.write_to(&mut cursor, image::ImageFormat::Jpeg)
+        .map_err(|e| format!("Failed to encode image as JPEG: {e}"))?;
+    let b64 = BASE64.encode(&jpeg_buf);
+    Ok(format!("data:image/jpeg;base64,{b64}"))
+}
+
+/// Expand 1-bit-per-pixel data to 8-bit grayscale.
+fn expand_1bit_to_8bit(data: &[u8], width: u32, height: u32) -> Vec<u8> {
+    let row_bytes = (width as usize).div_ceil(8);
+    let mut out = Vec::with_capacity((width * height) as usize);
+    for y in 0..height as usize {
+        for x in 0..width as usize {
+            let byte_idx = y * row_bytes + x / 8;
+            let bit_idx = 7 - (x % 8);
+            let bit = if byte_idx < data.len() {
+                (data[byte_idx] >> bit_idx) & 1
+            } else {
+                0
+            };
+            // In PDF, 0 is typically black for 1-bit images
+            out.push(if bit == 0 { 0 } else { 255 });
+        }
+    }
+    out
+}
+
+// ---------------------------------------------------------------------------
+// TIFF wrapper for CCITT Group 3/4 data
+// ---------------------------------------------------------------------------
+
+/// Wrap raw CCITT Group 3/4 data in a minimal TIFF container so the `image`
+/// crate (via its `tiff` backend) can decode it.
+fn wrap_ccitt_as_tiff(ccitt_data: &[u8], width: u32, height: u32, k: i32) -> Vec<u8> {
+    // TIFF compression tag: 3 = CCITT Group 3, 4 = CCITT Group 4
+    let compression: u16 = if k < 0 { 4 } else { 3 };
+
+    let num_entries: u16 = 8;
+    let ifd_size = 2 + (num_entries as u32) * 12 + 4;
+    let data_offset: u32 = 8 + ifd_size;
+
+    let mut buf = Vec::with_capacity(data_offset as usize + ccitt_data.len());
+
+    // TIFF header (8 bytes)
+    buf.extend_from_slice(b"II"); // little-endian
+    buf.extend_from_slice(&42u16.to_le_bytes()); // magic
+    buf.extend_from_slice(&8u32.to_le_bytes()); // offset to first IFD
+
+    // IFD entry count
+    buf.extend_from_slice(&num_entries.to_le_bytes());
+
+    // IFD entries (must be sorted by tag)
+    write_ifd_long(&mut buf, 256, width); // ImageWidth
+    write_ifd_long(&mut buf, 257, height); // ImageLength
+    write_ifd_short(&mut buf, 258, 1); // BitsPerSample
+    write_ifd_short(&mut buf, 259, compression); // Compression
+    write_ifd_short(&mut buf, 262, 0); // PhotometricInterpretation (WhiteIsZero)
+    write_ifd_long(&mut buf, 273, data_offset); // StripOffsets
+    write_ifd_long(&mut buf, 278, height); // RowsPerStrip
+    write_ifd_long(&mut buf, 279, ccitt_data.len() as u32); // StripByteCounts
+
+    // Next IFD offset (0 = no more IFDs)
+    buf.extend_from_slice(&0u32.to_le_bytes());
+
+    // Image data
+    buf.extend_from_slice(ccitt_data);
+
+    buf
+}
+
+fn write_ifd_short(buf: &mut Vec<u8>, tag: u16, value: u16) {
+    buf.extend_from_slice(&tag.to_le_bytes());
+    buf.extend_from_slice(&3u16.to_le_bytes()); // type = SHORT
+    buf.extend_from_slice(&1u32.to_le_bytes()); // count
+    buf.extend_from_slice(&(value as u32).to_le_bytes());
+}
+
+fn write_ifd_long(buf: &mut Vec<u8>, tag: u16, value: u32) {
+    buf.extend_from_slice(&tag.to_le_bytes());
+    buf.extend_from_slice(&4u16.to_le_bytes()); // type = LONG
+    buf.extend_from_slice(&1u32.to_le_bytes()); // count
+    buf.extend_from_slice(&value.to_le_bytes());
+}
+
+// ---------------------------------------------------------------------------
+// Helpers
+// ---------------------------------------------------------------------------
+
+/// Read an integer value from a PDF stream dictionary.
+fn get_stream_int(dict: &lopdf::Dictionary, key: &[u8]) -> Option<i64> {
+    dict.get(key).ok().and_then(|v| v.as_i64().ok())
+}
+
+/// Read an integer from the DecodeParms sub-dictionary of a stream.
+fn get_decode_parm_int(dict: &lopdf::Dictionary, key: &[u8]) -> Option<i64> {
+    dict.get(b"DecodeParms")
+        .ok()
+        .and_then(|v| v.as_dict().ok())
+        .and_then(|d| d.get(key).ok())
+        .and_then(|v| v.as_i64().ok())
 }
 
 #[cfg(test)]
 mod tests {
-    use super::extract_document_content;
+    use super::*;
     use base64::{engine::general_purpose::STANDARD as BASE64, Engine};
 
     #[tokio::test]
@@ -118,4 +467,121 @@ mod tests {
             "unexpected error: {err}"
         );
     }
+
+    #[tokio::test]
+    async fn extract_bitcoin_whitepaper_pdf() {
+        // Read the Bitcoin whitepaper PDF from the test fixtures directory (not embedded in binary)
+        let pdf_path = std::path::Path::new(env!("CARGO_MANIFEST_DIR"))
+            .join("test_fixtures/bitcoin_whitepaper.pdf");
+        let pdf_bytes = std::fs::read(&pdf_path).unwrap_or_else(|e| {
+            panic!("failed to read test fixture at {}: {e}", pdf_path.display())
+        });
+
+        let file_base64 = BASE64.encode(&pdf_bytes);
+
+        let resp =
+            extract_document_content(file_base64, "bitcoin.pdf".to_string(), "pdf".to_string())
+                .await
+                .expect("expected Bitcoin PDF extraction to succeed");
+
+        assert_eq!(resp.status, "completed");
+        assert_eq!(resp.document.filename, "bitcoin.pdf");
+
+        let content = &resp.document.text_content;
+
+        // Verify meaningful content was extracted (whitepaper is ~9 pages)
+        assert!(
+            content.len() > 1000,
+            "expected substantial content from Bitcoin whitepaper, got {} chars",
+            content.len()
+        );
+
+        // Verify key content from the Bitcoin whitepaper is present
+        assert!(
+            content.contains("Bitcoin") || content.contains("bitcoin"),
+            "expected 'Bitcoin' in extracted text"
+        );
+        assert!(
+            content.contains("peer-to-peer") || content.contains("peer to peer"),
+            "expected 'peer-to-peer' in extracted text"
+        );
+        assert!(
+            content.contains("Satoshi") || content.contains("Nakamoto"),
+            "expected author name in extracted text"
+        );
+    }
+
+    #[tokio::test]
+    async fn extract_scanned_pdf_returns_page_images() {
+        // Read a real scanned PDF (a scanned letter) from the test fixtures directory.
+        // Since it's a scanned document, pdf-inspector won't extract text -- instead
+        // we should get page images back for vision-model OCR.
+        let pdf_path = std::path::Path::new(env!("CARGO_MANIFEST_DIR"))
+            .join("test_fixtures/scanned_letter.pdf");
+        let pdf_bytes = std::fs::read(&pdf_path).unwrap_or_else(|e| {
+            panic!("failed to read test fixture at {}: {e}", pdf_path.display())
+        });
+
+        let file_base64 = BASE64.encode(&pdf_bytes);
+
+        let resp = extract_document_content(
+            file_base64,
+            "scanned_letter.pdf".to_string(),
+            "pdf".to_string(),
+        )
+        .await
+        .expect("expected scanned PDF extraction to succeed (with page images)");
+
+        assert_eq!(resp.status, "completed");
+        assert_eq!(resp.document.filename, "scanned_letter.pdf");
+
+        // Text content should be empty for a scanned PDF
+        assert!(
+            resp.document.text_content.is_empty(),
+            "scanned PDF should have empty text_content"
+        );
+
+        // Should have page images for OCR
+        let page_images = resp
+            .document
+            .page_images
+            .expect("scanned PDF should have page_images");
+
+        assert_eq!(
+            page_images.len(),
+            1,
+            "scanned letter has 1 page, expected 1 image"
+        );
+
+        // Each image should be a valid JPEG data URL
+        assert!(
+            page_images[0].starts_with("data:image/jpeg;base64,"),
+            "page image should be a JPEG data URL"
+        );
+
+        // Verify the image has reasonable size (not empty/corrupt)
+        let b64_data = page_images[0]
+            .strip_prefix("data:image/jpeg;base64,")
+            .unwrap();
+        let decoded_size = BASE64.decode(b64_data).unwrap().len();
+        assert!(
+            decoded_size > 1000,
+            "JPEG image should be >1KB, got {} bytes",
+            decoded_size
+        );
+    }
+
+    #[test]
+    fn test_wrap_ccitt_as_tiff_structure() {
+        let fake_data = vec![0u8; 100];
+        let tiff = wrap_ccitt_as_tiff(&fake_data, 100, 200, -1);
+
+        // Check TIFF header
+        assert_eq!(&tiff[0..2], b"II"); // little-endian
+        assert_eq!(u16::from_le_bytes([tiff[2], tiff[3]]), 42); // magic
+        assert_eq!(u32::from_le_bytes([tiff[4], tiff[5], tiff[6], tiff[7]]), 8); // IFD offset
+
+        // Should end with our image data
+        assert_eq!(&tiff[tiff.len() - 100..], &fake_data[..]);
+    }
 }
diff --git a/frontend/src-tauri/test_fixtures/bitcoin_whitepaper.pdf b/frontend/src-tauri/test_fixtures/bitcoin_whitepaper.pdf
new file mode 100644
index 00000000..1e19b739
Binary files /dev/null and b/frontend/src-tauri/test_fixtures/bitcoin_whitepaper.pdf differ
diff --git a/frontend/src-tauri/test_fixtures/scanned_letter.pdf b/frontend/src-tauri/test_fixtures/scanned_letter.pdf
new file mode 100644
index 00000000..8d3068e3
Binary files /dev/null and b/frontend/src-tauri/test_fixtures/scanned_letter.pdf differ
diff --git a/frontend/src/components/UnifiedChat.tsx b/frontend/src/components/UnifiedChat.tsx
index 830958ee..757db7cf 100644
--- a/frontend/src/components/UnifiedChat.tsx
+++ b/frontend/src/components/UnifiedChat.tsx
@@ -1733,6 +1733,7 @@ export function UnifiedChat() {
             document: {
               filename: string;
               text_content: string;
+              page_images?: string[];
             };
             status: string;
           }
@@ -1743,7 +1744,23 @@ export function UnifiedChat() {
             fileType: "pdf"
           });
 
-          if (result.document?.text_content) {
+          if (result.document?.page_images?.length) {
+            // Scanned/image-based PDF — store placeholder only (no base64 image data).
+            // page_images are intentionally excluded from documentText to avoid
+            // sending multi-MB base64 blobs as plain text to the AI model.
+            // Vision-model OCR integration will use a separate code path.
+            const scannedData = {
+              document: {
+                filename: result.document.filename,
+                text_content:
+                  "[Scanned PDF: " +
+                  result.document.page_images.length +
+                  " page image(s) extracted. OCR via vision model is not yet supported.]"
+              }
+            };
+            setDocumentText(JSON.stringify(scannedData));
+            setDocumentName(file.name);
+          } else if (result.document?.text_content) {
             // Create a cleaned version with image references removed
             const cleanedParsed = {
               document: {