fix:修正子agent无法正确接收本地图片(参考图)路径的问题

astrbot/core/astr_agent_tool_exec.py

@@ -1,9 +1,12 @@
 import asyncio
 import inspect
 import json
+import os
 import traceback
 import typing as T
 import uuid
+from collections.abc import Sequence
+from collections.abc import Set as AbstractSet
 
 import mcp
 
@@ -26,6 +29,7 @@
     SEND_MESSAGE_TO_USER_TOOL,
 )
 from astrbot.core.cron.events import CronMessageEvent
+from astrbot.core.message.components import Image
 from astrbot.core.message.message_event_result import (
     CommandResult,
     MessageChain,
@@ -35,9 +39,110 @@
 from astrbot.core.provider.entites import ProviderRequest
 from astrbot.core.provider.register import llm_tools
 from astrbot.core.utils.history_saver import persist_agent_history
+from astrbot.core.utils.string_utils import normalize_and_dedupe_strings
 
 
 class FunctionToolExecutor(BaseFunctionToolExecutor[AstrAgentContext]):
+    _ALLOWED_IMAGE_EXTENSIONS = {
+        ".png",
+        ".jpg",
+        ".jpeg",
+        ".gif",
+        ".webp",
+        ".bmp",
+        ".tif",
+        ".tiff",
+        ".svg",
+        ".heic",
+    }
+
+    @classmethod
+    def _is_supported_image_ref(cls, image_ref: str) -> bool:
+        if not image_ref:
+            return False
+        lowered = image_ref.lower()
+        if lowered.startswith(("http://", "https://", "base64://")):
+            return True
+        file_path = image_ref[8:] if lowered.startswith("file:///") else image_ref
+        ext = os.path.splitext(file_path)[1].lower()
+        if ext in cls._ALLOWED_IMAGE_EXTENSIONS:
+            return True
+        # Keep support for extension-less temp files returned by image converters.
+        return ext == "" and os.path.exists(file_path)
+
+    @classmethod
+    def _collect_image_urls_from_args(cls, image_urls_raw: T.Any) -> list[str]:
+        candidates: list[str] = []
+        if image_urls_raw is None:
+            pass
+        elif isinstance(image_urls_raw, str):
+            candidates.append(image_urls_raw)
+        elif isinstance(image_urls_raw, (Sequence, AbstractSet)) and not isinstance(
+            image_urls_raw, (str, bytes, bytearray)
+        ):
+            non_string_count = 0
+            for item in image_urls_raw:
+                if isinstance(item, str):
+                    candidates.append(item)
+                else:
+                    non_string_count += 1
+            if non_string_count > 0:
+                logger.warning(
+                    "Dropped %d non-string image_urls entries in handoff tool args.",
+                    non_string_count,
+                )
+        else:
+            logger.warning(
+                "Unsupported image_urls type in handoff tool args: %s",
+                type(image_urls_raw).__name__,
+            )
+        return candidates
+
+    @classmethod
+    async def _collect_image_urls_from_message(
+        cls, run_context: ContextWrapper[AstrAgentContext]
+    ) -> list[str]:
+        urls: list[str] = []
+        event = getattr(run_context.context, "event", None)
+        message_obj = getattr(event, "message_obj", None)
+        message = getattr(message_obj, "message", None)
+        if message:
+            for idx, component in enumerate(message):
+                if not isinstance(component, Image):
+                    continue
+                try:
+                    path = await component.convert_to_file_path()
+                    if path:
+                        urls.append(path)
+                except Exception as e:
+                    logger.error(
+                        "Failed to convert handoff image component at index %d: %s",
+                        idx,
+                        e,
+                        exc_info=True,
+                    )
+        return urls
+
+    @classmethod
+    async def _collect_handoff_image_urls(
+        cls,
+        run_context: ContextWrapper[AstrAgentContext],
+        image_urls_raw: T.Any,
+    ) -> list[str]:
+        candidates: list[str] = []
+        candidates.extend(cls._collect_image_urls_from_args(image_urls_raw))
+        candidates.extend(await cls._collect_image_urls_from_message(run_context))
+
+        normalized = normalize_and_dedupe_strings(candidates)
+        sanitized = [item for item in normalized if cls._is_supported_image_ref(item)]
+        dropped_count = len(normalized) - len(sanitized)
+        if dropped_count > 0:
+            logger.warning(
+                "Dropped %d invalid image_urls entries in handoff tool args.",
+                dropped_count,
+            )
+        return sanitized
+
     @classmethod
     async def execute(cls, tool, run_context, **tool_args):
         """执行函数调用。
@@ -58,7 +163,7 @@ async def execute(cls, tool, run_context, **tool_args):
                 ):
                     yield r
                 return
-            async for r in cls._execute_handoff(tool, run_context, **tool_args):
+            async for r in cls._execute_handoff(tool, run_context, tool_args):
                 yield r
             return
 
@@ -161,10 +266,14 @@ async def _execute_handoff(
         cls,
         tool: HandoffTool,
         run_context: ContextWrapper[AstrAgentContext],
-        **tool_args,
+        tool_args: dict[str, T.Any],
     ):
         input_ = tool_args.get("input")
-        image_urls = tool_args.get("image_urls")
+        image_urls = await cls._collect_handoff_image_urls(
+            run_context,
+            tool_args.get("image_urls"),
+        )
+        tool_args["image_urls"] = image_urls
 
         # Build handoff toolset from registered tools plus runtime computer tools.
         toolset = cls._build_handoff_toolset(run_context, tool.agent.tools)
@@ -264,7 +373,7 @@ async def _do_handoff_background(
         """Run the subagent handoff and, on completion, wake the main agent."""
         result_text = ""
         try:
-            async for r in cls._execute_handoff(tool, run_context, **tool_args):
+            async for r in cls._execute_handoff(tool, run_context, tool_args):
                 if isinstance(r, mcp.types.CallToolResult):
                     for content in r.content:
                         if isinstance(content, mcp.types.TextContent):

tests/unit/test_astr_agent_tool_exec.py

@@ -0,0 +1,135 @@
+from types import SimpleNamespace
+
+import mcp
+import pytest
+
+from astrbot.core.agent.run_context import ContextWrapper
+from astrbot.core.astr_agent_tool_exec import FunctionToolExecutor
+from astrbot.core.message.components import Image
+
+
+class _DummyEvent:
+    def __init__(self, message_components: list[object] | None = None) -> None:
+        self.unified_msg_origin = "webchat:FriendMessage:webchat!user!session"
+        self.message_obj = SimpleNamespace(message=message_components or [])
+
+    def get_extra(self, _key: str):
+        return None
+
+
+class _DummyTool:
+    def __init__(self) -> None:
+        self.name = "transfer_to_subagent"
+        self.agent = SimpleNamespace(name="subagent")
+
+
+def _build_run_context(message_components: list[object] | None = None):
+    event = _DummyEvent(message_components=message_components)
+    ctx = SimpleNamespace(event=event, context=SimpleNamespace())
+    return ContextWrapper(context=ctx)
+
+
+@pytest.mark.asyncio
+async def test_collect_handoff_image_urls_normalizes_filters_and_appends_event_image(
+    monkeypatch: pytest.MonkeyPatch,
+):
+    async def _fake_convert_to_file_path(self):
+        return "/tmp/event_image.png"
+
+    monkeypatch.setattr(Image, "convert_to_file_path", _fake_convert_to_file_path)
+
+    run_context = _build_run_context([Image(file="file:///tmp/original.png")])
+    image_urls_input = (
+        " https://example.com/a.png ",
+        "/tmp/not_an_image.txt",
+        "/tmp/local.webp",
+        123,
+    )
+
+    image_urls = await FunctionToolExecutor._collect_handoff_image_urls(
+        run_context,
+        image_urls_input,
+    )
+
+    assert image_urls == [
+        "https://example.com/a.png",
+        "/tmp/local.webp",
+        "/tmp/event_image.png",
+    ]
+
+
+@pytest.mark.asyncio
+async def test_collect_handoff_image_urls_skips_failed_event_image_conversion(
+    monkeypatch: pytest.MonkeyPatch,
+):
+    async def _fake_convert_to_file_path(self):
+        raise RuntimeError("boom")
+
+    monkeypatch.setattr(Image, "convert_to_file_path", _fake_convert_to_file_path)
+
+    run_context = _build_run_context([Image(file="file:///tmp/original.png")])
+    image_urls = await FunctionToolExecutor._collect_handoff_image_urls(
+        run_context,
+        ["https://example.com/a.png"],
+    )
+
+    assert image_urls == ["https://example.com/a.png"]
+
+
+@pytest.mark.asyncio
+async def test_do_handoff_background_reports_prepared_image_urls(
+    monkeypatch: pytest.MonkeyPatch,
+):
+    captured: dict = {}
+
+    async def _fake_execute_handoff(cls, tool, run_context, tool_args):
+        tool_args["image_urls"] = ["https://example.com/raw.png"]
+        yield mcp.types.CallToolResult(
+            content=[mcp.types.TextContent(type="text", text="ok")]
+        )
+
+    async def _fake_wake(cls, run_context, **kwargs):
+        captured.update(kwargs)
+
+    monkeypatch.setattr(
+        FunctionToolExecutor,
+        "_execute_handoff",
+        classmethod(_fake_execute_handoff),
+    )
+    monkeypatch.setattr(
+        FunctionToolExecutor,
+        "_wake_main_agent_for_background_result",
+        classmethod(_fake_wake),
+    )
+
+    run_context = _build_run_context()
+    await FunctionToolExecutor._do_handoff_background(
+        tool=_DummyTool(),
+        run_context=run_context,
+        task_id="task-id",
+        input="hello",
+        image_urls="https://example.com/raw.png",
+    )
+
+    assert captured["tool_args"]["image_urls"] == ["https://example.com/raw.png"]
+
+
+@pytest.mark.asyncio
+async def test_collect_handoff_image_urls_keeps_extensionless_existing_event_file(
+    monkeypatch: pytest.MonkeyPatch,
+):
+    async def _fake_convert_to_file_path(self):
+        return "/tmp/astrbot-handoff-image"
+
+    monkeypatch.setattr(Image, "convert_to_file_path", _fake_convert_to_file_path)
+    monkeypatch.setattr(
+        "astrbot.core.astr_agent_tool_exec.os.path.exists", lambda _: True
+    )
+
+    run_context = _build_run_context([Image(file="file:///tmp/original.png")])
+    image_urls = await FunctionToolExecutor._collect_handoff_image_urls(
+        run_context,
+        [],
+    )
+
+    assert image_urls == ["/tmp/astrbot-handoff-image"]