picoclaw/pkg/tools/fs/load_image.go

164 lines
4.6 KiB
Go
Raw Normal View History

package fstools
feat: add load_image tool for local file vision (#2116) * feat: add load_image tool for local file vision * fix: address load_image PR review feedback - Exclude load_image from sub-agent tools via Unregister after Clone, since RunToolLoop does not call resolveMediaRefs - Add ToolRegistry.Unregister() method - Fix scope collision: use channel:chatID instead of filename - Add channel/chatID context resolution matching send_file pattern - Add comment explaining iteration > 1 guard on resolveMediaRefs - Remove emoji from ForUser for consistency with send_file - Add load_image_test.go * feat: enable load_image for subagents via MediaResolver in RunToolLoop Instead of removing load_image from sub-agent tools (28f69e71), inject a MediaResolver into the legacy RunToolLoop fallback path so media:// refs are resolved to base64 before each LLM call — matching the main agent loop behavior. - Add MediaResolver field to ToolLoopConfig and call it on iteration > 1 - Add SubagentManager.SetMediaResolver() and wire it through runTask - Remove ToolRegistry.Unregister() (no longer needed) - Restore load_image in sub-agent tool set (revert Clone+Unregister) - Add TestSubagentManager_SetMediaResolver_StoresResolver * refactor(load_image): remove prompt parameter from tool schema * test(tools): add success-path test for LoadImageTool Add TestLoadImage_SuccessPath that creates a real PNG file with valid magic bytes, calls Execute with WithToolContext, and verifies: - result.IsError == false - ToolResult.Media contains a media:// ref - ToolResult.ForLLM contains the [image: marker - media ref is resolvable in the store Add explanatory comment in loop.go for why Media and ArtifactTags coexist on non-ResponseHandled tool results (e.g. load_image). * fix: preallocate slice in tests and add ResponseHandled guard in toolloop Fix prealloc linter failure in load_image_test.go. Prevent double-resolving media by checking ResponseHandled in toolloop.go. * Register TTS tool if provider is available --------- Co-authored-by: Reusu <admin@yumao.name> Co-authored-by: 美電球 <hoshina@evaz.org>
2026-04-01 13:32:10 +00:00
import (
"context"
"fmt"
"os"
"path/filepath"
"regexp"
"strings"
"github.com/sipeed/picoclaw/pkg/config"
"github.com/sipeed/picoclaw/pkg/media"
)
// LoadImageTool loads a local image file into the MediaStore and returns a
// media:// reference. The agent loop's resolveMediaRefs will then base64-encode
// it and attach it as an image_url part in the next LLM request, enabling
// vision on local files — the same pipeline used when a user sends an image
// through a chat channel.
//
// This is intentionally different from SendFileTool:
// - SendFileTool → MediaResult + WithResponseHandled() → sends file to user, ends turn
// - LoadImageTool → plain ToolResult with media:// in ForLLM → LLM sees the image next turn
type LoadImageTool struct {
workspace string
restrict bool
maxFileSize int
mediaStore media.MediaStore
allowPaths []*regexp.Regexp
defaultChannel string
defaultChatID string
}
func NewLoadImageTool(
workspace string,
restrict bool,
maxFileSize int,
store media.MediaStore,
allowPaths ...[]*regexp.Regexp,
) *LoadImageTool {
if maxFileSize <= 0 {
maxFileSize = config.DefaultMaxMediaSize
}
var patterns []*regexp.Regexp
if len(allowPaths) > 0 {
patterns = allowPaths[0]
}
return &LoadImageTool{
workspace: workspace,
restrict: restrict,
maxFileSize: maxFileSize,
mediaStore: store,
allowPaths: patterns,
}
}
func (t *LoadImageTool) Name() string { return "load_image" }
func (t *LoadImageTool) Description() string {
return "Load a local image file so you can analyze its contents with vision. " +
"Supported formats: JPEG, PNG, GIF, WebP, BMP. " +
"After calling this tool, describe or analyze the image in your next response."
}
func (t *LoadImageTool) Parameters() map[string]any {
return map[string]any{
"type": "object",
"properties": map[string]any{
"path": map[string]any{
"type": "string",
"description": "Path to the local image file. Relative paths are resolved from workspace.",
},
},
"required": []string{"path"},
}
}
func (t *LoadImageTool) SetContext(channel, chatID string) {
t.defaultChannel = channel
t.defaultChatID = chatID
}
func (t *LoadImageTool) SetMediaStore(store media.MediaStore) {
t.mediaStore = store
}
func (t *LoadImageTool) Execute(ctx context.Context, args map[string]any) *ToolResult {
path, _ := args["path"].(string)
if strings.TrimSpace(path) == "" {
return ErrorResult("path is required")
}
// Prefer context-injected channel/chatID (set by ExecuteWithContext), fall back to SetContext values.
channel := ToolChannel(ctx)
if channel == "" {
channel = t.defaultChannel
}
chatID := ToolChatID(ctx)
if chatID == "" {
chatID = t.defaultChatID
}
if channel == "" || chatID == "" {
return ErrorResult("no target channel/chat available")
}
if t.mediaStore == nil {
return ErrorResult("media store not configured")
}
resolved, err := validatePathWithAllowPaths(path, t.workspace, t.restrict, t.allowPaths)
if err != nil {
return ErrorResult(fmt.Sprintf("invalid path: %v", err))
}
info, err := os.Stat(resolved)
if err != nil {
return ErrorResult(fmt.Sprintf("file not found: %v", err))
}
if info.IsDir() {
return ErrorResult("path is a directory, expected an image file")
}
if info.Size() > int64(t.maxFileSize) {
return ErrorResult(fmt.Sprintf(
"file too large: %d bytes (max %d bytes)", info.Size(), t.maxFileSize,
))
}
// Detect MIME type — reuse the helper already in send_file.go
mediaType := detectMediaType(resolved)
if !strings.HasPrefix(mediaType, "image/") {
return ErrorResult(fmt.Sprintf(
"file does not appear to be an image (detected type: %s)", mediaType,
))
}
filename := filepath.Base(resolved)
scope := fmt.Sprintf("tool:load_image:%s:%s", channel, chatID)
ref, err := t.mediaStore.Store(resolved, media.MediaMeta{
Filename: filename,
ContentType: mediaType,
Source: "tool:load_image",
CleanupPolicy: media.CleanupPolicyForgetOnly,
}, scope)
if err != nil {
return ErrorResult(fmt.Sprintf("failed to register image in media store: %v", err))
}
fix(feishu): fix image download with API fallback and post image support (#2708) * fix(feishu): fix image download with API fallback and post image support - Add Image.Get API fallback when MessageResource.Get fails (different permission scope: im:resource vs im:message:readonly) - Extract and download images from post (rich text) messages - Extract images from interactive card messages - Deduplicate post image keys across locales - Add comprehensive tests for new helpers Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * feat(media): add image path tags alongside base64 for LLM file access Images are still base64-encoded into msg.Media for multimodal LLMs, but now also get [image:path] tags injected into message content so the LLM knows the local file path for save/forward operations. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * refactor(media): only auto-inject images for tool results, not user messages Channel-received images (role=user) now get path tags only, letting the LLM decide whether to view via load_image or just operate on the file. Tool result images (role=tool, e.g. load_image) are base64-encoded into a synthetic user message appended after the tool message, since many LLM APIs don't support image_url in tool messages. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix(media): preserve tool-message ordering for multi-tool-call scenarios Move synthetic user message (carrying base64 tool images) to after the entire contiguous tool-message block instead of immediately after each tool message. This preserves the assistant→tool→tool ordering required by OpenAI-compatible APIs. Also fix load_image to use generic [image: photo] placeholder so injectPathTags can properly replace it with the actual path. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix(test): update load_image test for [image: photo] placeholder The test was checking ForLLM for the media:// ref, but load_image now emits the generic [image: photo] placeholder instead. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix(media): match all channel image placeholders in injectPathTags Different channels emit different placeholder formats — Telegram/Feishu use [image: photo], WeCom/WeChat/Line use bare [image], QQ/Discord use [image: <filename>]. The previous string-match code only handled [image: photo], so for the other channels the path tag was appended as a duplicate, producing content like "[image] [image:/path]". Switch to per-type regex that matches all generic placeholder shapes while leaving path tags ([image:/path]) untouched. Also fixes the same issue for [audio], [video], [file] tags. Added test coverage for the various placeholder shapes. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix(media): skip path tag append for JSON content (Feishu cards/posts) When content is structured JSON (interactive cards, post messages), injectPathTags now skips the fallback append — only placeholder replacement is attempted. This prevents corrupting JSON payloads like {"schema":"2.0",...} with appended [image:/path] tags. Adds looksLikeJSON() helper and three test cases covering JSON objects, arrays, and an end-to-end resolveMediaRefs scenario with Feishu card content. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(media): prepend path tags for JSON content, narrow looksLikeJSON Two fixes from code review: 1. looksLikeJSON now only checks for '{' prefix (not '['), avoiding false positives on regular text like "[update] see attached". 2. For JSON content (Feishu cards/posts), path tags are prepended before the JSON instead of being silently dropped. This ensures the LLM can discover attached images via the path tag while the JSON payload stays valid for downstream parsing. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
2026-04-30 03:08:00 +00:00
// Build the tool result text. The media:// ref in Media will be picked
// up by resolveMediaRefs in agent_media.go and base64-encoded for tool
// result messages (role="tool"), so the LLM can see the image content.
msg := fmt.Sprintf("Image loaded: %s\n[image: photo]", filename)
feat: add load_image tool for local file vision (#2116) * feat: add load_image tool for local file vision * fix: address load_image PR review feedback - Exclude load_image from sub-agent tools via Unregister after Clone, since RunToolLoop does not call resolveMediaRefs - Add ToolRegistry.Unregister() method - Fix scope collision: use channel:chatID instead of filename - Add channel/chatID context resolution matching send_file pattern - Add comment explaining iteration > 1 guard on resolveMediaRefs - Remove emoji from ForUser for consistency with send_file - Add load_image_test.go * feat: enable load_image for subagents via MediaResolver in RunToolLoop Instead of removing load_image from sub-agent tools (28f69e71), inject a MediaResolver into the legacy RunToolLoop fallback path so media:// refs are resolved to base64 before each LLM call — matching the main agent loop behavior. - Add MediaResolver field to ToolLoopConfig and call it on iteration > 1 - Add SubagentManager.SetMediaResolver() and wire it through runTask - Remove ToolRegistry.Unregister() (no longer needed) - Restore load_image in sub-agent tool set (revert Clone+Unregister) - Add TestSubagentManager_SetMediaResolver_StoresResolver * refactor(load_image): remove prompt parameter from tool schema * test(tools): add success-path test for LoadImageTool Add TestLoadImage_SuccessPath that creates a real PNG file with valid magic bytes, calls Execute with WithToolContext, and verifies: - result.IsError == false - ToolResult.Media contains a media:// ref - ToolResult.ForLLM contains the [image: marker - media ref is resolvable in the store Add explanatory comment in loop.go for why Media and ArtifactTags coexist on non-ResponseHandled tool results (e.g. load_image). * fix: preallocate slice in tests and add ResponseHandled guard in toolloop Fix prealloc linter failure in load_image_test.go. Prevent double-resolving media by checking ResponseHandled in toolloop.go. * Register TTS tool if provider is available --------- Co-authored-by: Reusu <admin@yumao.name> Co-authored-by: 美電球 <hoshina@evaz.org>
2026-04-01 13:32:10 +00:00
return &ToolResult{
ForLLM: msg,
ForUser: fmt.Sprintf("Loaded image: %s", filename),
// Media refs inside ForLLM are resolved by resolveMediaRefs in the
// agent loop before the next LLM call. Do NOT use MediaResult here —
// that would send the file to the user channel instead.
Media: []string{ref},
}
}