feat(tools/io): 内置 IO/网络工具（file(路径安全)/http(SSRF防护)/web-search/web-fetch/knowledge）+ 大小截断

plugins/tools/io/web_fetch.go

@@ -0,0 +1,109 @@
+package io
+
+import (
+	"context"
+	"fmt"
+	"log/slog"
+	"mime"
+	"net/http"
+	"strings"
+
+	"git.yeij.top/AskaEth/Cyrene_v2/pkg/plugins/sdk"
+)
+
+// Compile-time assertion that webFetchTool implements sdk.Tool.
+var _ sdk.Tool = (*webFetchTool)(nil)
+
+// webFetchTool fetches a URL over HTTP and returns its readable text, limited by
+// size and content type, extracting text from HTML when appropriate.
+type webFetchTool struct {
+	client       *http.Client
+	allowedHosts []string
+	maxBytes     int64
+	log          *slog.Logger
+}
+
+func newWebFetchTool(t *tools) *webFetchTool {
+	return &webFetchTool{
+		client:       t.client,
+		allowedHosts: t.allowedHosts,
+		maxBytes:     t.maxFetchBytes,
+		log:          t.log,
+	}
+}
+
+// Name implements sdk.Tool.
+func (w *webFetchTool) Name() string { return "web-fetch" }
+
+// Description implements sdk.Tool.
+func (w *webFetchTool) Description() string {
+	return "Fetch a URL and return its readable text, limited by size and content type."
+}
+
+// Parameters implements sdk.Tool.
+func (w *webFetchTool) Parameters() (map[string]any, error) {
+	return map[string]any{
+		"type": "object",
+		"properties": map[string]any{
+			"url":       map[string]any{"type": "string", "description": "Absolute http(s) URL."},
+			"max_bytes": map[string]any{"type": "integer", "description": "Response size cap in bytes."},
+		},
+		"required": []any{"url"},
+	}, nil
+}
+
+// Execute implements sdk.Tool.
+func (w *webFetchTool) Execute(ctx context.Context, args any) (any, error) {
+	m, ok := args.(map[string]any)
+	if !ok {
+		return nil, fmt.Errorf("io: web-fetch: args must be an object")
+	}
+	rawURL := argString(m, "url")
+	if strings.TrimSpace(rawURL) == "" {
+		return nil, fmt.Errorf("io: web-fetch: url is required")
+	}
+	if _, err := validateRequestURL(ctx, rawURL, w.allowedHosts); err != nil {
+		return nil, err
+	}
+	if w.client == nil {
+		return nil, fmt.Errorf("io: web-fetch: no underlying client")
+	}
+	req, err := http.NewRequestWithContext(ctx, http.MethodGet, rawURL, nil)
+	if err != nil {
+		return nil, fmt.Errorf("io: web-fetch: build request: %w", err)
+	}
+	resp, err := w.client.Do(req)
+	if err != nil {
+		return nil, fmt.Errorf("io: web-fetch: do: %w", err)
+	}
+	defer resp.Body.Close()
+	if resp.StatusCode != http.StatusOK {
+		return nil, fmt.Errorf("io: web-fetch: url returned status %d", resp.StatusCode)
+	}
+	mt := "text/plain"
+	if rawCT := resp.Header.Get("Content-Type"); rawCT != "" {
+		var err error
+		mt, _, err = mime.ParseMediaType(rawCT)
+		if err != nil || !isTextContentType(mt) {
+			return nil, fmt.Errorf("io: web-fetch: unexpected content type %q", rawCT)
+		}
+	}
+	max := w.maxBytes
+	if v, ok := argInt(m, "max_bytes"); ok && v > 0 {
+		max = int64(v)
+	}
+	data, truncated, err := readLimited(resp.Body, max)
+	if err != nil {
+		return nil, fmt.Errorf("io: web-fetch: read: %w", err)
+	}
+	text := string(data)
+	if isHTML(mt) {
+		text = htmlToText(text)
+	}
+	return map[string]any{
+		"url":          rawURL,
+		"content_type": mt,
+		"text":         text,
+		"truncated":    truncated,
+	}, nil
+}