botrun-flow-lang 5.12.264__py3-none-any.whl → 6.2.61__py3-none-any.whl

botrun_flow_lang/mcp_server/default_mcp.py

@@ -34,7 +34,6 @@ from botrun_flow_lang.utils.botrun_logger import get_default_botrun_logger
 
 # Import for generate_image
 from langchain_community.utilities.dalle_image_generator import DallEAPIWrapper
-from langchain_community.callbacks import get_openai_callback
 
 # Initialize MCP server
 mcp = FastMCP(name="BotrunFlowLangDefaultMCP", stateless_http=True)
@@ -74,7 +73,7 @@ async def scrape(url: str) -> dict:
 @mcp.tool()
 async def chat_with_pdf(
     pdf_url: str, user_input: str, botrun_flow_lang_url: str, user_id: str
-) -> str:
+) -> dict:
     """
     Analyze a PDF file and answer questions about its content.
 
@@ -89,7 +88,16 @@ async def chat_with_pdf(
         user_id: REQUIRED - User ID for file upload (LLM can get this from system prompt)
 
     Returns:
-        str: Analysis result or Plotly-compatible data structure if visualization is needed
+        dict: {
+            "result": str,  # Analysis result
+            "usage_metadata": List[Dict]  # Token usage for each LLM call, with format:
+                - prompt_tokens: int
+                - completion_tokens: int
+                - total_tokens: int
+                - cache_creation_input_tokens: int
+                - cache_read_input_tokens: int
+                - model: str
+        }
     """
     logger.info(f"chat_with_pdf pdf_url: {pdf_url} user_input: {user_input}")
 
@@ -106,7 +114,7 @@ async def chat_with_imgs(
     user_input: str,
     botrun_flow_lang_url: str,
     user_id: str,
-) -> str:
+) -> dict:
     """
     Analyze multiple images and answer questions about their content.
 
@@ -117,7 +125,16 @@ async def chat_with_imgs(
         user_id: REQUIRED - User ID for file upload (LLM can get this from system prompt)
 
     Returns:
-        str: Analysis result or Plotly-compatible data structure if visualization is needed
+        dict: {
+            "result": str,  # Analysis result
+            "usage_metadata": List[Dict]  # Token usage for each LLM call, with format:
+                - prompt_tokens: int
+                - completion_tokens: int
+                - total_tokens: int
+                - cache_creation_input_tokens: int
+                - cache_read_input_tokens: int
+                - model: str
+        }
     """
     logger.info(f"chat_with_imgs img_urls: {img_urls} user_input: {user_input}")
 
@@ -136,7 +153,7 @@ async def chat_with_imgs(
 @mcp.tool()
 async def generate_image(
     user_input: str, user_id: str = "", botrun_flow_lang_url: str = ""
-) -> str:
+) -> dict:
     """
     Generate high-quality images using DALL-E 3 and store permanently in GCS.
 
@@ -173,18 +190,35 @@ async def generate_image(
         botrun_flow_lang_url: REQUIRED - URL for the botrun flow lang API (LLM can get this from system prompt)
 
     Returns:
-        str: Permanent URL to the generated image stored in GCS, or error message if generation fails
+        dict: {
+            "result": str,  # Permanent URL to the generated image stored in GCS, or error message
+            "usage_metadata": List[Dict]  # Token usage for each LLM call, with format:
+                - prompt_tokens: int
+                - completion_tokens: int
+                - total_tokens: int
+                - cache_creation_input_tokens: int
+                - cache_read_input_tokens: int
+                - model: str
+        }
     """
+    usage_list = []
+
     try:
         logger.info(f"generate_image user_input: {user_input}")
 
         # 驗證必要參數
         if not user_id:
             logger.error("User ID not available")
-            return "User ID not available"
+            return {
+                "result": "User ID not available",
+                "usage_metadata": usage_list,
+            }
         if not botrun_flow_lang_url:
             logger.error("botrun_flow_lang_url not available")
-            return "botrun_flow_lang_url not available"
+            return {
+                "result": "botrun_flow_lang_url not available",
+                "usage_metadata": usage_list,
+            }
 
         # Check rate limit before generating image
         rate_limit_client = RateLimitClient()
@@ -201,26 +235,38 @@ async def generate_image(
                 f"User {user_id} has reached daily limit of {daily_limit} image generations. "
                 f"Current usage: {current_usage}. Please try again tomorrow."
             )
-            return f"[Please tell user error] You have reached your daily limit of {daily_limit} image generations. " \
-                f"Current usage: {current_usage}. Please try again tomorrow."
-            # raise BotrunRateLimitException(
-            #     f"You have reached your daily limit of {daily_limit} image generations. "
-            #     f"Current usage: {current_usage}. Please try again tomorrow."
-            # )
+            return {
+                "result": f"[Please tell user error] You have reached your daily limit of {daily_limit} image generations. "
+                    f"Current usage: {current_usage}. Please try again tomorrow.",
+                "usage_metadata": usage_list,
+            }
 
         # 2. 使用 DALL-E 生成圖片
+        dalle_size = "1024x1024"  # 可選: 1024x1024, 1024x1792, 1792x1024
+        dalle_quality = "standard"  # 可選: standard, hd
         dalle_wrapper = DallEAPIWrapper(
-            api_key=os.getenv("OPENAI_API_KEY"), model="dall-e-3"
+            api_key=os.getenv("OPENAI_API_KEY"),
+            model="dall-e-3",
+            size=dalle_size,
+            quality=dalle_quality,
         )
 
-        # Generate image with token usage tracking
-        with get_openai_callback() as cb:
-            temp_image_url = dalle_wrapper.run(user_input)
-            logger.info(
-                f"DALL-E generated temporary URL: {temp_image_url}, "
-                f"prompt tokens: {cb.prompt_tokens}, "
-                f"completion tokens: {cb.completion_tokens}"
-            )
+        # Generate image (DALL-E charges per image, not per token)
+        temp_image_url = dalle_wrapper.run(user_input)
+        logger.info(f"DALL-E generated temporary URL: {temp_image_url}")
+
+        # DALL-E 不使用 token 計費，改記錄圖片生成次數和規格
+        usage_list.append({
+            "prompt_tokens": 0,
+            "completion_tokens": 0,
+            "total_tokens": 0,
+            "cache_creation_input_tokens": 0,
+            "cache_read_input_tokens": 0,
+            "model": "dall-e-3",
+            "image_count": 1,
+            "image_size": dalle_size,
+            "image_quality": dalle_quality,
+        })
 
         # 3. 下載並上傳到 GCS，取得永久 URL
         from botrun_flow_lang.langgraph_agents.agents.util.local_files import (
@@ -236,22 +282,34 @@ async def generate_image(
             # 4. 更新使用計數
             await rate_limit_client.update_drawing_usage(user_id)
 
-            return permanent_url
+            return {
+                "result": permanent_url,
+                "usage_metadata": usage_list,
+            }
         except Exception as upload_error:
             logger.error(
                 f"Failed to upload to GCS, returning temporary URL: {upload_error}"
             )
             # Fallback: 回傳臨時 URL
             await rate_limit_client.update_drawing_usage(user_id)
-            return temp_image_url
+            return {
+                "result": temp_image_url,
+                "usage_metadata": usage_list,
+            }
 
     except Exception as e:
         logger.error(f"generate_image error: {e}", error=str(e), exc_info=True)
 
         # Check if this is a user-visible exception
         if str(e).startswith("[Please tell user error]"):
-            return str(e)  # Return the error message as is
-        return f"Error: {e}"
+            return {
+                "result": str(e),
+                "usage_metadata": usage_list,
+            }
+        return {
+            "result": f"Error: {e}",
+            "usage_metadata": usage_list,
+        }
 
 
 @mcp.tool()
@@ -267,7 +325,7 @@ async def generate_tmp_public_url(
         user_id: REQUIRED - User ID for file upload (LLM can get this from system prompt)
 
     Returns:
-        str: A public URL that can be used to access the file for 7 days
+        str: A temporary public URL that may be deleted periodically
 
     Raises:
         FileNotFoundError: If the specified file does not exist
@@ -289,7 +347,9 @@ async def create_html_page(
     user_id: str,
 ) -> str:
     """
-    Create a custom HTML page and return its URL.
+    Create a custom HTML page and return a PERMANENT URL that never expires.
+
+    The URL created by this tool will remain accessible indefinitely.
 
     This tool supports complete HTML documents, including JavaScript and CSS, which can be used to create
     complex interactive pages.
@@ -314,8 +374,9 @@ async def create_html_page(
         user_id: REQUIRED - User ID for file upload (LLM can get this from system prompt)
 
     Returns:
-        str: URL for the HTML page. This URL should be provided to the user,
-             as they will need to access it to view the content in their web browser.
+        str: A permanent URL for the HTML page that never expires.
+             This URL should be provided to the user, as they will need to
+             access it to view the content in their web browser.
     """
     try:
         logger.info(f"create_html_page html_content: {html_content} title: {title}")
@@ -692,11 +753,23 @@ async def web_search(
         user_id: Optional user ID (not used for this tool)
 
     Returns:
-        dict: A dictionary containing:
-              - content (str): The detailed answer based on web search results
-              - citations (list): A list of URLs, citations are important to provide to the user
-              - images (list): A list of image URLs (only when return_images is True)
+        dict: {
+            "result": {
+                "content": str,  # The detailed answer based on web search results
+                "citations": list,  # A list of URLs
+                "images": list  # A list of image URLs (only when return_images is True)
+            },
+            "usage_metadata": List[Dict]  # Token usage for each LLM call, with format:
+                - prompt_tokens: int
+                - completion_tokens: int
+                - total_tokens: int
+                - cache_creation_input_tokens: int
+                - cache_read_input_tokens: int
+                - model: str
+        }
     """
+    usage_list = []
+
     try:
         logger.info(f"web_search user_input: {user_input}")
 
@@ -717,6 +790,7 @@ async def web_search(
             "content": "",
             "citations": [],
         }
+        raw_response = None
 
         async for event in respond_with_perplexity_search(
             final_input,
@@ -729,16 +803,52 @@ async def web_search(
         ):
             if event and isinstance(event.chunk, str):
                 search_result = json.loads(event.chunk)
+            if event and event.raw_json:
+                raw_response = event.raw_json
+
+        # Extract usage from raw response
+        if raw_response and "usage" in raw_response:
+            usage = raw_response["usage"]
+            prompt_tokens = usage.get("prompt_tokens", 0) or 0
+            citation_tokens = usage.get("citation_tokens", 0) or 0
+            completion_tokens = usage.get("completion_tokens", 0) or 0
+            model = raw_response.get("model", "sonar-reasoning-pro")
+
+            # 判斷是否使用 OpenRouter（與 perplexity_search.py 邏輯一致）
+            is_use_openrouter = os.getenv("OPENROUTER_API_KEY") and os.getenv("OPENROUTER_BASE_URL")
+            if return_images:
+                is_use_openrouter = False
+
+            # 加上 API 來源前綴
+            if is_use_openrouter:
+                model = f"openrouter/{model}"
+            else:
+                # 直接使用 Perplexity API，加上 perplexity/ 前綴
+                if not model.startswith("perplexity/"):
+                    model = f"perplexity/{model}"
+
+            usage_list.append({
+                "prompt_tokens": prompt_tokens + citation_tokens,
+                "completion_tokens": completion_tokens,
+                "total_tokens": prompt_tokens + citation_tokens + completion_tokens,
+                "cache_creation_input_tokens": 0,
+                "cache_read_input_tokens": 0,
+                "model": model,
+            })
 
         logger.info(
             f"web_search completed============> {len(search_result.get('content', ''))}"
         )
-        return (
-            search_result
-            if search_result
-            else {"content": "No results found.", "citations": []}
-        )
+
+        result = search_result if search_result else {"content": "No results found.", "citations": []}
+        return {
+            "result": result,
+            "usage_metadata": usage_list,
+        }
 
     except Exception as e:
         logger.error(f"web_search error: {e}", error=str(e), exc_info=True)
-        return {"content": f"Error during web search: {str(e)}", "citations": []}
+        return {
+            "result": {"content": f"Error during web search: {str(e)}", "citations": []},
+            "usage_metadata": usage_list,
+        }

{botrun_flow_lang-5.12.264.dist-info → botrun_flow_lang-6.2.61.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: botrun-flow-lang
-Version: 5.12.264
+Version: 6.2.61
 Summary: A flow language for botrun
 Author-email: sebastian-hsu <sebastian.hsu@gmail.com>
 License: MIT
@@ -27,12 +27,11 @@ Requires-Dist: google-cloud-storage<3,>=2.18
 Requires-Dist: google-genai>=1.28.0
 Requires-Dist: jinja2>=3.1.6
 Requires-Dist: langchain-anthropic>=0.3.10
-Requires-Dist: langchain-aws>=0.2.17
+Requires-Dist: langchain-aws>=1.0.0
 Requires-Dist: langchain-community>=0.3.27
-Requires-Dist: langchain-core>=0.3.72
+Requires-Dist: langchain-core>=1.1.2
 Requires-Dist: langchain-google-community>=2.0.3
-Requires-Dist: langchain-google-genai>=2.0.9
-Requires-Dist: langchain-google-vertexai<3.0.0,>=2.1.2
+Requires-Dist: langchain-google-genai>=4.0.0
 Requires-Dist: langchain-mcp-adapters>=0.1.7
 Requires-Dist: langchain-openai>=0.3.28
 Requires-Dist: langchain>=0.3.27
@@ -41,7 +40,7 @@ Requires-Dist: langgraph-supervisor>=0.0.20
 Requires-Dist: langgraph>=0.6.3
 Requires-Dist: line-bot-sdk>=3.17.1
 Requires-Dist: mcp<1.11.0,>=1.10.1
-Requires-Dist: numpy<2,>=1
+Requires-Dist: numpy>=1.24.0
 Requires-Dist: openai>=1.99.1
 Requires-Dist: pandas>=2.2.3
 Requires-Dist: pdfminer-six==20250506

{botrun_flow_lang-5.12.264.dist-info → botrun_flow_lang-6.2.61.dist-info}/RECORD

@@ -9,7 +9,8 @@ botrun_flow_lang/api/auth_utils.py,sha256=KoVTZUMOBaATWvdyjjYKdBjDu8MaQGGvmhE8gn
 botrun_flow_lang/api/botrun_back_api.py,sha256=qNIQqMFZ969XaLE4qsbM659bcoK11o9sy85gfmeO8Kw,2462
 botrun_flow_lang/api/flow_api.py,sha256=I6ZMohJOpuVcs8q2euUjdydz0xYvavRei7f3LQFmjbQ,111
 botrun_flow_lang/api/hatch_api.py,sha256=S-_bNt4Y8oKtlzXW7JA3TuMRFO-Pb4-5OobhnwfiqFE,17492
-botrun_flow_lang/api/langgraph_api.py,sha256=PP0K_H5-BRJsjFGIvZEZubLtQ97FXp35Ts8SGSKxKe8,30102
+botrun_flow_lang/api/langgraph_api.py,sha256=E1FDme6CUnMRXRaQFhp_S-uI4m7vtQo5CZOd0o_X1nA,30228
+botrun_flow_lang/api/langgraph_constants.py,sha256=oxh3Rj940mZ7ekKIiQodvpQs_pek_R0atqgda9yxSV0,411
 botrun_flow_lang/api/line_bot_api.py,sha256=JluAbySIU42zWc0NaMwL1fhfRCEGMjAVWTfwKXp2F0A,56984
 botrun_flow_lang/api/model_api.py,sha256=vkzVvzxxsAhqbiMcVAeqiQoheJVbPLAXBqwoU5PgWMw,9783
 botrun_flow_lang/api/rate_limit_api.py,sha256=zrQ9wFILNqYMiLDM8NqdfcDg87BdyzbBC2Kns89WIGo,980
@@ -24,7 +25,7 @@ botrun_flow_lang/api/youtube_api.py,sha256=9eGr--gR2OoM9JZ6Nf9KqPiE-FeXEx8R-QeJv
 botrun_flow_lang/langgraph_agents/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 botrun_flow_lang/langgraph_agents/agents/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 botrun_flow_lang/langgraph_agents/agents/agent_runner.py,sha256=tiuPIqAcM8rIWBTjo8NS4owTepCsX3QkIHaUEDakOTc,6673
-botrun_flow_lang/langgraph_agents/agents/langgraph_react_agent.py,sha256=ANXWs6WKRULX9eECipCj_ivY-mnyt-gcu9_-xmLGKSg,30672
+botrun_flow_lang/langgraph_agents/agents/langgraph_react_agent.py,sha256=-4ejdM9SiocZyV92wCg1DOKNMBUTO6-wanj-D73RG7k,31156
 botrun_flow_lang/langgraph_agents/agents/search_agent_graph.py,sha256=hWDPt0U09Gj-3-NNWhsn9xaakYbOcHExIXqcL8TeZxw,32046
 botrun_flow_lang/langgraph_agents/agents/agent_tools/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 botrun_flow_lang/langgraph_agents/agents/agent_tools/step_planner.py,sha256=S4TYt0ZhgdAZ-2ndH8hJoEaIyDKdNJdWHjEZ49Lg_NQ,2427
@@ -38,18 +39,20 @@ botrun_flow_lang/langgraph_agents/agents/gov_researcher/gov_researcher_graph.py,
 botrun_flow_lang/langgraph_agents/agents/tools/__init__.py,sha256=-z1uuC3IET02q8kPhPlr-L9eTGJqgHjEJlC__cG16H0,105
 botrun_flow_lang/langgraph_agents/agents/tools/gemini_code_execution.py,sha256=EEp8xhVU-Kj1Nk5qV8ObqdVZ8gT6GITrE4VyjIc2InA,14238
 botrun_flow_lang/langgraph_agents/agents/util/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+botrun_flow_lang/langgraph_agents/agents/util/custom_vertex_claude.py,sha256=RDj1-PnoR36xdOUu7r6yMQrxAe-CXyGcSZEKiOlPxJY,14951
 botrun_flow_lang/langgraph_agents/agents/util/gemini_grounding.py,sha256=JTfH9WJNDlpvMvfzXyZy3bHeCN58MTnEOiamQGMsqh0,2884
 botrun_flow_lang/langgraph_agents/agents/util/html_util.py,sha256=g5yJO0qTqRq_kb-xhSnWX3WAbHDIjNQYl7ErRBPQwHs,13230
-botrun_flow_lang/langgraph_agents/agents/util/img_util.py,sha256=px_ymI5U8WkFujqF-nbxZZNFOKPxJyj5HYV2IQdpYl4,10405
+botrun_flow_lang/langgraph_agents/agents/util/img_util.py,sha256=6OERtpGGimlev4Pb_O1UbMNaT_DMBHSmAgo9gB-R8xk,12385
 botrun_flow_lang/langgraph_agents/agents/util/local_files.py,sha256=b7N4B3P9zPPDj7_C9y8JaU5oROQostCXBt7wfxi_L64,13529
 botrun_flow_lang/langgraph_agents/agents/util/mermaid_util.py,sha256=o65I979wA7jzzB_Zp-t2CxBjNdXyFtdTkdofi4bJlb0,2642
 botrun_flow_lang/langgraph_agents/agents/util/model_utils.py,sha256=oeYEwiEtlrNGomKZ98M3F_OvXYjAIoCV9IJCY9eMuug,4954
-botrun_flow_lang/langgraph_agents/agents/util/pdf_analyzer.py,sha256=Q8cuTYvYG5ZSCdjPFiH96dWjXrSXA9DSZpLicKMx_u0,16177
+botrun_flow_lang/langgraph_agents/agents/util/pdf_analyzer.py,sha256=Pu5hZGquvgm1Iy6qWD5SDG2--9tumTHdJ8EGLcTF8LU,20305
 botrun_flow_lang/langgraph_agents/agents/util/pdf_cache.py,sha256=nmHA-qf2px3ywUtC_5kXIQHg-Gl2W8DOSnL8gOFR3xY,7375
 botrun_flow_lang/langgraph_agents/agents/util/pdf_processor.py,sha256=1YbB4zpabQB-8HwRvd4LRyye4oSusLNUW_iJfmUNANw,6244
 botrun_flow_lang/langgraph_agents/agents/util/perplexity_search.py,sha256=dZmb4tSECEXWOSZkqdMhFrmnGwMhdHSUJvLT7IAVu_s,19537
 botrun_flow_lang/langgraph_agents/agents/util/plotly_util.py,sha256=WwAPcmMDnQnrsxH_92377G0yRWf-dF-g8uOG9KnkcCk,1972
 botrun_flow_lang/langgraph_agents/agents/util/tavily_search.py,sha256=jjuA8dko9YRSs_LcvMduAsSGDaix3UEzw4cIllVVFh0,6822
+botrun_flow_lang/langgraph_agents/agents/util/usage_metadata.py,sha256=ahjulhlNUdyLG-KNBL-0pQPkxbzpiYKjGR2YqQCF1fA,1207
 botrun_flow_lang/langgraph_agents/agents/util/youtube_util.py,sha256=PHGDpJqRZNHLoLMAFpfpQiz_vlZWG3u53GZQajjjEpI,3007
 botrun_flow_lang/langgraph_agents/cache/__init__.py,sha256=SnKEKUXeTReKzUeNVXfvP3BEZypgKBQ4TKs_-T8ZdtI,36
 botrun_flow_lang/langgraph_agents/cache/langgraph_botrun_cache.py,sha256=HgwP7HKIglm24LFy7sIddLNi-nAfmIkxqvkwV6FnxVk,6364
@@ -58,7 +61,7 @@ botrun_flow_lang/llm_agent/llm_agent.py,sha256=1yPws2a5MVouOBnsNudh-c1POviYaz9hn
 botrun_flow_lang/llm_agent/llm_agent_util.py,sha256=cvnkHYH1D1V1_chgIByCb1Cn7iytNxtlJpfrFlYa_a4,3131
 botrun_flow_lang/log/.gitignore,sha256=ZeCRrK8PsUdGyHBMDfCkk1Jl9XrN9VkgJmyeCIUCxGU,18
 botrun_flow_lang/mcp_server/__init__.py,sha256=lbhwcb-QsYmdXA8bS3pSD-CLVbcbCKfl1XeOaUm380Y,218
-botrun_flow_lang/mcp_server/default_mcp.py,sha256=wNzK0WRS62ABFb5f0QuJQMDpNs4yZSreGTzmD8fwXqs,29234
+botrun_flow_lang/mcp_server/default_mcp.py,sha256=azFgYEo0yXwBm-n-SqFwSw2YhcrvgAloS_YNR8eXsO0,33194
 botrun_flow_lang/models/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 botrun_flow_lang/models/token_usage.py,sha256=PHLPwzLaGwFhWzxaHBYcahztlyTZEpRso5XI6pscWVM,876
 botrun_flow_lang/models/nodes/utils.py,sha256=uKCdPYQfzjGf8Bzoy-FSZxtcwVifwnCviaJM4qnEyrI,6904
@@ -97,6 +100,6 @@ botrun_flow_lang/utils/yaml_utils.py,sha256=dPlabIol-Clhnwc7N5nuffCaLSq8dyvmvjRw
 botrun_flow_lang/utils/clients/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 botrun_flow_lang/utils/clients/rate_limit_client.py,sha256=96NNCHB9I5C5bpVFF6sfPhmh4oAx3UdOLb-Z4PAXLdg,8558
 botrun_flow_lang/utils/clients/token_verify_client.py,sha256=-AnYApJ9CvxVn-RhCCZZ2LCrf065fgskhwLKAm-aiN0,5893
-botrun_flow_lang-5.12.264.dist-info/METADATA,sha256=u_ZfQqG71uK5H-aV7BNSjYAHTB5MF6kjW0sh4VOioD0,6221
-botrun_flow_lang-5.12.264.dist-info/WHEEL,sha256=WLgqFyCfm_KASv4WHyYy0P3pM_m7J5L9k2skdKLirC8,87
-botrun_flow_lang-5.12.264.dist-info/RECORD,,
+botrun_flow_lang-6.2.61.dist-info/METADATA,sha256=7mfsw0_BTSzWq5fSeKqCg0RmuOEyMfdKwItnzlRA0p8,6164
+botrun_flow_lang-6.2.61.dist-info/WHEEL,sha256=WLgqFyCfm_KASv4WHyYy0P3pM_m7J5L9k2skdKLirC8,87
+botrun_flow_lang-6.2.61.dist-info/RECORD,,