diff --git a/go.mod b/go.mod
index ac288e56d..520cb2b1c 100644
--- a/go.mod
+++ b/go.mod
@@ -22,13 +22,14 @@ require (
 	github.com/open-dingtalk/dingtalk-stream-sdk-go v0.9.1
 	github.com/openai/openai-go/v3 v3.22.0
 	github.com/rivo/tview v0.42.0
+	github.com/rs/zerolog v1.34.0
 	github.com/slack-go/slack v0.17.3
 	github.com/spf13/cobra v1.10.2
 	github.com/stretchr/testify v1.11.1
 	github.com/tencent-connect/botgo v0.2.1
 	go.mau.fi/whatsmeow v0.0.0-20260219150138-7ae702b1eed4
 	golang.org/x/oauth2 v0.36.0
-	golang.org/x/term v0.40.0
+	golang.org/x/term v0.41.0
 	golang.org/x/time v0.14.0
 	google.golang.org/protobuf v1.36.11
 	gopkg.in/yaml.v3 v3.0.1
@@ -54,7 +55,6 @@ require (
 	github.com/pmezard/go-difflib v1.0.0 // indirect
 	github.com/remyoudompheng/bigfft v0.0.0-20230129092748-24d4a6f8daec // indirect
 	github.com/rivo/uniseg v0.4.7 // indirect
-	github.com/rs/zerolog v1.34.0 // indirect
 	github.com/segmentio/asm v1.1.3 // indirect
 	github.com/segmentio/encoding v0.5.3 // indirect
 	github.com/spf13/pflag v1.0.10 // indirect
@@ -62,7 +62,7 @@ require (
 	go.mau.fi/libsignal v0.2.1 // indirect
 	go.mau.fi/util v0.9.6 // indirect
 	golang.org/x/exp v0.0.0-20260212183809-81e46e3db34a // indirect
-	golang.org/x/text v0.34.0 // indirect
+	golang.org/x/text v0.35.0 // indirect
 	modernc.org/libc v1.67.6 // indirect
 	modernc.org/mathutil v1.7.1 // indirect
 	modernc.org/memory v1.11.0 // indirect
@@ -92,8 +92,8 @@ require (
 	github.com/valyala/fastjson v1.6.10 // indirect
 	github.com/yosida95/uritemplate/v3 v3.0.2 // indirect
 	golang.org/x/arch v0.24.0 // indirect
-	golang.org/x/crypto v0.48.0
-	golang.org/x/net v0.51.0 // indirect
-	golang.org/x/sync v0.19.0 // indirect
-	golang.org/x/sys v0.41.0 // indirect
+	golang.org/x/crypto v0.49.0
+	golang.org/x/net v0.51.0
+	golang.org/x/sync v0.20.0 // indirect
+	golang.org/x/sys v0.42.0 // indirect
 )
diff --git a/go.sum b/go.sum
index f0e3fc132..25d941bc7 100644
--- a/go.sum
+++ b/go.sum
@@ -250,8 +250,8 @@ golang.org/x/crypto v0.0.0-20200622213623-75b288015ac9/go.mod h1:LzIPMQfyMNhhGPh
 golang.org/x/crypto v0.0.0-20210421170649-83a5a9bb288b/go.mod h1:T9bdIzuCu7OtxOm1hfPfRQxPLYneinmdGuTeoZ9dtd4=
 golang.org/x/crypto v0.0.0-20210921155107-089bfa567519/go.mod h1:GvvjBRRGRdwPK5ydBHafDWAxML/pGHZbMvKqRZ5+Abc=
 golang.org/x/crypto v0.16.0/go.mod h1:gCAAfMLgwOJRpTjQ2zCCt2OcSfYMTeZVSRtQlPC7Nq4=
-golang.org/x/crypto v0.48.0 h1:/VRzVqiRSggnhY7gNRxPauEQ5Drw9haKdM0jqfcCFts=
-golang.org/x/crypto v0.48.0/go.mod h1:r0kV5h3qnFPlQnBSrULhlsRfryS2pmewsg+XfMgkVos=
+golang.org/x/crypto v0.49.0 h1:+Ng2ULVvLHnJ/ZFEq4KdcDd/cfjrrjjNSXNzxg0Y4U4=
+golang.org/x/crypto v0.49.0/go.mod h1:ErX4dUh2UM+CFYiXZRTcMpEcN8b/1gxEuv3nODoYtCA=
 golang.org/x/exp v0.0.0-20260212183809-81e46e3db34a h1:ovFr6Z0MNmU7nH8VaX5xqw+05ST2uO1exVfZPVqRC5o=
 golang.org/x/exp v0.0.0-20260212183809-81e46e3db34a/go.mod h1:K79w1Vqn7PoiZn+TkNpx3BUWUQksGO3JcVX6qIjytmA=
 golang.org/x/mod v0.2.0/go.mod h1:s0Qsj1ACt9ePp/hMypM3fl4fZqREWJwdYDEqhRiZZUA=
@@ -284,8 +284,8 @@ golang.org/x/sync v0.0.0-20190911185100-cd5d95a43a6e/go.mod h1:RxMgew5VJxzue5/jJ
 golang.org/x/sync v0.0.0-20201020160332-67f06af15bc9/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sync v0.0.0-20220722155255-886fb9371eb4/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sync v0.1.0/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
-golang.org/x/sync v0.19.0 h1:vV+1eWNmZ5geRlYjzm2adRgW2/mcpevXNg50YZtPCE4=
-golang.org/x/sync v0.19.0/go.mod h1:9KTHXmSnoGruLpwFjVSX0lNNA75CykiMECbovNTZqGI=
+golang.org/x/sync v0.20.0 h1:e0PTpb7pjO8GAtTs2dQ6jYa5BWYlMuX047Dco/pItO4=
+golang.org/x/sync v0.20.0/go.mod h1:9xrNwdLfx4jkKbNva9FpL6vEN7evnE43NNNJQ2LF3+0=
 golang.org/x/sys v0.0.0-20180909124046-d0be0721c37e/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
 golang.org/x/sys v0.0.0-20190215142949-d0b11bdaac8a/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
 golang.org/x/sys v0.0.0-20190412213103-97732733099d/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
@@ -307,15 +307,15 @@ golang.org/x/sys v0.6.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.8.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.12.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.15.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
-golang.org/x/sys v0.41.0 h1:Ivj+2Cp/ylzLiEU89QhWblYnOE9zerudt9Ftecq2C6k=
-golang.org/x/sys v0.41.0/go.mod h1:OgkHotnGiDImocRcuBABYBEXf8A9a87e/uXjp9XT3ks=
+golang.org/x/sys v0.42.0 h1:omrd2nAlyT5ESRdCLYdm3+fMfNFE/+Rf4bDIQImRJeo=
+golang.org/x/sys v0.42.0/go.mod h1:4GL1E5IUh+htKOUEOaiffhrAeqysfVGipDYzABqnCmw=
 golang.org/x/term v0.0.0-20201126162022-7de9c90e9dd1/go.mod h1:bj7SfCRtBDWHUb9snDiAeCFNEtKQo2Wmx5Cou7ajbmo=
 golang.org/x/term v0.0.0-20210927222741-03fcf44c2211/go.mod h1:jbD1KX2456YbFQfuXm/mYQcufACuNUgVhRMnK/tPxf8=
 golang.org/x/term v0.5.0/go.mod h1:jMB1sMXY+tzblOD4FWmEbocvup2/aLOaQEp7JmGp78k=
 golang.org/x/term v0.8.0/go.mod h1:xPskH00ivmX89bAKVGSKKtLOWNx2+17Eiy94tnKShWo=
 golang.org/x/term v0.15.0/go.mod h1:BDl952bC7+uMoWR75FIrCDx79TPU9oHkTZ9yRbYOrX0=
-golang.org/x/term v0.40.0 h1:36e4zGLqU4yhjlmxEaagx2KuYbJq3EwY8K943ZsHcvg=
-golang.org/x/term v0.40.0/go.mod h1:w2P8uVp06p2iyKKuvXIm7N/y0UCRt3UfJTfZ7oOpglM=
+golang.org/x/term v0.41.0 h1:QCgPso/Q3RTJx2Th4bDLqML4W6iJiaXFq2/ftQF13YU=
+golang.org/x/term v0.41.0/go.mod h1:3pfBgksrReYfZ5lvYM0kSO0LIkAl4Yl2bXOkKP7Ec2A=
 golang.org/x/text v0.3.0/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=
 golang.org/x/text v0.3.3/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=
 golang.org/x/text v0.3.6/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=
@@ -323,8 +323,8 @@ golang.org/x/text v0.3.7/go.mod h1:u+2+/6zg+i71rQMx5EYifcz6MCKuco9NR6JIITiCfzQ=
 golang.org/x/text v0.7.0/go.mod h1:mrYo+phRRbMaCq/xk9113O4dZlRixOauAjOtrjsXDZ8=
 golang.org/x/text v0.9.0/go.mod h1:e1OnstbJyHTd6l/uOt8jFFHp6TRDWZR/bV3emEE/zU8=
 golang.org/x/text v0.14.0/go.mod h1:18ZOQIKpY8NJVqYksKHtTdi31H5itFRjB5/qKTNYzSU=
-golang.org/x/text v0.34.0 h1:oL/Qq0Kdaqxa1KbNeMKwQq0reLCCaFtqu2eNuSeNHbk=
-golang.org/x/text v0.34.0/go.mod h1:homfLqTYRFyVYemLBFl5GgL/DWEiH5wcsQ5gSh1yziA=
+golang.org/x/text v0.35.0 h1:JOVx6vVDFokkpaq1AEptVzLTpDe9KGpj5tR4/X+ybL8=
+golang.org/x/text v0.35.0/go.mod h1:khi/HExzZJ2pGnjenulevKNX1W67CUy0AsXcNubPGCA=
 golang.org/x/time v0.14.0 h1:MRx4UaLrDotUKUdCIqzPC48t1Y9hANFKIRpNx+Te8PI=
 golang.org/x/time v0.14.0/go.mod h1:eL/Oa2bBBK0TkX57Fyni+NgnyQQN4LitPmob2Hjnqw4=
 golang.org/x/tools v0.0.0-20180917221912-90fa682c2a6e/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
diff --git a/pkg/agent/loop_media.go b/pkg/agent/loop_media.go
index 6b0a211f4..12750c2d7 100644
--- a/pkg/agent/loop_media.go
+++ b/pkg/agent/loop_media.go
@@ -525,16 +525,31 @@ func (al *AgentLoop) ocrPDF(
 	}
 	hash := mediacache.HashData(hashInput)
 
-	// Check cache
+	// Check cache (both text extraction and OCR)
 	if al.mediaCache != nil {
+		if entry, ok := al.mediaCache.GetEntry(hash, mediacache.TypePDFText); ok {
+			logger.DebugCF("agent", "PDF text cache hit", map[string]any{"hash": hash})
+			return formatDocumentTag(entry.Result, entry.FilePath, entry.Pages)
+		}
 		if entry, ok := al.mediaCache.GetEntry(hash, mediacache.TypePDFOCR); ok {
 			logger.DebugCF("agent", "PDF OCR cache hit", map[string]any{"hash": hash})
 			return formatDocumentTag(entry.Result, entry.FilePath, entry.Pages)
 		}
 	}
 
-	// Get page count for progress display
-	totalPages := mediacache.PDFPageCount(pdfPath)
+	// Fast path: try pdftotext for PDFs with a text layer (skip if figures requested).
+	// pdftotext is orders of magnitude faster than OCR.
+	if !withFigures {
+		if text, pages, ok := tryPdftotextExtract(ctx, pdfPath); ok {
+			return al.savePdftotextResult(pdfPath, text, pages, hash)
+		}
+	}
+
+	// Get page count for progress display (use pdfinfo, fall back to regex)
+	totalPages := pdfinfoPageCount(ctx, pdfPath)
+	if totalPages == 0 {
+		totalPages = mediacache.PDFPageCount(pdfPath)
+	}
 	totalStr := mediacache.FormatPageCount(totalPages)
 
 	// Send hint message (only if not already sent by Phase 1 waitForPDFFollowUp)
diff --git a/pkg/agent/loop_media_pypdf.go b/pkg/agent/loop_media_pypdf.go
new file mode 100644
index 000000000..589da2db3
--- /dev/null
+++ b/pkg/agent/loop_media_pypdf.go
@@ -0,0 +1,141 @@
+package agent
+
+import (
+	"context"
+	"fmt"
+	"os"
+	"os/exec"
+	"path/filepath"
+	"strconv"
+	"strings"
+	"time"
+	"unicode"
+
+	"github.com/sipeed/picoclaw/pkg/logger"
+	"github.com/sipeed/picoclaw/pkg/mediacache"
+)
+
+// minTextDensity is the minimum ratio of non-whitespace characters to total
+// runes for extracted text to be considered "real" (not garbage/metadata).
+const minTextDensity = 0.3
+
+// minTextLength is the minimum total rune count of extracted text for the
+// fast path to succeed. Very short texts likely mean the PDF is image-based.
+const minTextLength = 100
+
+// pdftotextTimeout is the maximum time for pdftotext text extraction.
+const pdftotextTimeout = 30 * time.Second
+
+// tryPdftotextExtract attempts to extract text from a PDF using pdftotext
+// (poppler-utils). Returns the extracted text, page count, and true if
+// successful. Returns "", 0, false if pdftotext is unavailable, PDF has
+// no text layer, or the extracted text is too short/garbage.
+func tryPdftotextExtract(ctx context.Context, pdfPath string) (text string, pages int, ok bool) {
+	cmdCtx, cancel := context.WithTimeout(ctx, pdftotextTimeout)
+	defer cancel()
+
+	// Extract text to stdout. -layout preserves reading order.
+	cmd := exec.CommandContext(cmdCtx, "pdftotext", "-layout", pdfPath, "-")
+	out, err := cmd.Output()
+	if err != nil {
+		logger.DebugCF("agent", "pdftotext extraction failed", map[string]any{
+			"path":  pdfPath,
+			"error": err.Error(),
+		})
+		return "", 0, false
+	}
+
+	extracted := strings.TrimSpace(string(out))
+
+	// Get page count via pdfinfo
+	pages = pdfinfoPageCount(ctx, pdfPath)
+
+	// Check quality: too short or too sparse → probably image-based PDF
+	runes := []rune(extracted)
+	if len(runes) < minTextLength {
+		logger.DebugCF("agent", "pdftotext: text too short, falling back to OCR", map[string]any{
+			"path":  pdfPath,
+			"runes": len(runes),
+		})
+		return "", pages, false
+	}
+
+	nonSpace := 0
+	for _, r := range runes {
+		if !unicode.IsSpace(r) {
+			nonSpace++
+		}
+	}
+	density := float64(nonSpace) / float64(len(runes))
+	if density < minTextDensity {
+		logger.DebugCF("agent", "pdftotext: low text density, falling back to OCR", map[string]any{
+			"path":    pdfPath,
+			"density": fmt.Sprintf("%.2f", density),
+		})
+		return "", pages, false
+	}
+
+	logger.InfoCF("agent", "pdftotext: text extraction successful", map[string]any{
+		"path":  pdfPath,
+		"pages": pages,
+		"runes": len(runes),
+	})
+
+	return extracted, pages, true
+}
+
+// pdfinfoPageCount runs `pdfinfo` to get the page count of a PDF.
+// Returns 0 if pdfinfo is unavailable or fails.
+func pdfinfoPageCount(ctx context.Context, pdfPath string) int {
+	cmdCtx, cancel := context.WithTimeout(ctx, 5*time.Second)
+	defer cancel()
+
+	cmd := exec.CommandContext(cmdCtx, "pdfinfo", pdfPath)
+	out, err := cmd.Output()
+	if err != nil {
+		return 0
+	}
+
+	for _, line := range strings.Split(string(out), "\n") {
+		if strings.HasPrefix(line, "Pages:") {
+			field := strings.TrimSpace(strings.TrimPrefix(line, "Pages:"))
+			if n, err := strconv.Atoi(field); err == nil {
+				return n
+			}
+		}
+	}
+	return 0
+}
+
+// savePdftotextResult writes extracted text to a .md file in the output
+// directory and caches it. Returns the document tag string.
+func (al *AgentLoop) savePdftotextResult(
+	pdfPath, text string, pages int, hash string,
+) string {
+	outputDir := al.ocrOutputDir()
+	os.MkdirAll(outputDir, 0o755)
+
+	base := strings.TrimSuffix(filepath.Base(pdfPath), filepath.Ext(pdfPath))
+	mdPath := filepath.Join(outputDir, base+"_text.md")
+
+	if err := os.WriteFile(mdPath, []byte(text), 0o644); err != nil {
+		logger.WarnCF("agent", "Failed to write pdftotext output", map[string]any{
+			"path":  mdPath,
+			"error": err.Error(),
+		})
+		return fmt.Sprintf("[file:%s]", pdfPath)
+	}
+
+	preview := extractPreview(text, maxPreviewRunes)
+
+	// Cache the result
+	if al.mediaCache != nil {
+		_ = al.mediaCache.PutEntry(hash, mediacache.TypePDFText, mediacache.Entry{
+			Result:   preview,
+			FilePath: mdPath,
+			Pages:    pages,
+		})
+	}
+
+	return formatDocumentTag(preview, mdPath, pages)
+}
diff --git a/pkg/mediacache/cache.go b/pkg/mediacache/cache.go
index 91439aedb..ecbfb3bb8 100644
--- a/pkg/mediacache/cache.go
+++ b/pkg/mediacache/cache.go
@@ -27,6 +27,7 @@ CREATE INDEX IF NOT EXISTS idx_media_cache_accessed ON media_cache(accessed_at);
 const (
 	TypeImageDesc = "image_desc"
 	TypePDFOCR    = "pdf_ocr"
+	TypePDFText   = "pdf_text" // pypdf text extraction (no OCR)
 )
 
 // Cache provides SQLite-backed caching for media processing results.