zrb 1.6.0__py3-none-any.whl → 1.6.1__py3-none-any.whl

zrb/llm_config.py

@@ -26,9 +26,10 @@ You are an expert in various fields including technology, science, history, and
 
 # Concise summarization focused on preserving critical context for continuity.
 DEFAULT_SUMMARIZATION_PROMPT = """
-You are a summarization assistant. Create an updated, concise summary integrating
-the previous summary (if any) with the new conversation history.
-Your primary goal is to preserve ALL critical context needed for the main assistant
+You are a summarization assistant.
+Your goal is to help main assistant to continue the conversation by creating an updated,
+concise summary integrating the previous summary (if any) with the new conversation history.
+Preserve ALL critical context needed for the main assistant
 to continue the task effectively. This includes key facts, decisions, tool usage
 results, and essential background. Do not omit details that would force the main
 assistant to re-gather information.
@@ -37,11 +38,13 @@ Output *only* the updated summary text.
 
 DEFAULT_CONTEXT_ENRICHMENT_PROMPT = """
 You are an information extraction assistant.
+Your goal is to help main assistant to continue the conversation by extracting
+important informations.
 Analyze the conversation history and current context to extract key facts like
 user_name, user_roles, preferences, goals, etc.
 Return only a JSON object containing a single key "response", whose value is
-another JSON object with these details.
-If nothing is found, output {"response": {}}.
+another JSON object with these details (i.e., {"response": {"context_name": "value"}}).
+If no context can be extracted, return {"response": {}}.
 """.strip()
 
 DEFAULT_SPECIAL_INSTRUCTION_PROMPT = ""  # Default to empty

zrb/task/llm/agent.py

@@ -2,14 +2,12 @@ from collections.abc import Callable
 from typing import TYPE_CHECKING, Any
 
 if TYPE_CHECKING:
-    from openai import APIError
     from pydantic_ai import Agent, Tool
     from pydantic_ai.agent import AgentRun
     from pydantic_ai.mcp import MCPServer
     from pydantic_ai.models import Model
     from pydantic_ai.settings import ModelSettings
 else:
-    APIError = Any
     Agent = Any
     Tool = Any
     AgentRun = Any
@@ -130,6 +128,7 @@ async def run_agent_iteration(
     Raises:
         Exception: If any error occurs during agent execution.
     """
+    from openai import APIError
     from pydantic_ai.messages import ModelMessagesTypeAdapter
 
     async with agent.run_mcp_servers():

zrb/task/llm/context.py

@@ -73,9 +73,7 @@ def get_conversation_context(
     Retrieves the conversation context.
     If a value in the context dict is callable, it executes it with ctx.
     """
-    raw_context = get_attr(
-        ctx, conversation_context_attr, {}, auto_render=False
-    )  # Context usually shouldn't be rendered
+    raw_context = get_attr(ctx, conversation_context_attr, {}, auto_render=False)
     if not isinstance(raw_context, dict):
         ctx.log_warning(
             f"Conversation context resolved to type {type(raw_context)}, "

zrb/task/llm/context_enrichment.py

@@ -1,6 +1,5 @@
 import json
 import traceback
-from textwrap import dedent
 from typing import TYPE_CHECKING, Any
 
 from pydantic import BaseModel
@@ -9,8 +8,13 @@ from zrb.attr.type import BoolAttr, IntAttr
 from zrb.context.any_context import AnyContext
 from zrb.llm_config import llm_config
 from zrb.task.llm.agent import run_agent_iteration
+from zrb.task.llm.history import (
+    count_part_in_history_list,
+    replace_system_prompt_in_history_list,
+)
 from zrb.task.llm.typing import ListOfDict
 from zrb.util.attr import get_bool_attr, get_int_attr
+from zrb.util.cli.style import stylize_faint
 
 if TYPE_CHECKING:
     from pydantic_ai.models import Model
@@ -47,15 +51,13 @@ async def enrich_context(
         context_json = json.dumps(conversation_context)
         history_json = json.dumps(history_list)
         # The user prompt will now contain the dynamic data
-        user_prompt_data = dedent(
-            f"""
-            Analyze the following
-            # Current Context
-            {context_json}
-            # Conversation History
-            {history_json}
-        """
-        ).strip()
+        user_prompt_data = "\n".join(
+            [
+                "Extract context from the following conversation info",
+                f"Existing Context: {context_json}",
+                f"Conversation History: {history_json}",
+            ]
+        )
     except Exception as e:
         ctx.log_warning(f"Error formatting context/history for enrichment: {e}")
         return conversation_context  # Return original context if formatting fails
@@ -72,6 +74,7 @@ async def enrich_context(
     )
 
     try:
+        ctx.print(stylize_faint("[Context Enrichment Triggered]"), plain=True)
         enrichment_run = await run_agent_iteration(
             ctx=ctx,
             agent=enrichment_agent,
@@ -80,6 +83,8 @@ async def enrich_context(
         )
         if enrichment_run and enrichment_run.result.output:
             response = enrichment_run.result.output.response
+            usage = enrichment_run.result.usage()
+            ctx.print(stylize_faint(f"[Token Usage] {usage}"), plain=True)
             if response:
                 conversation_context.update(response)
                 ctx.log_info("Context enriched based on history.")
@@ -127,15 +132,15 @@ def should_enrich_context(
     """
     Determines if context enrichment should occur based on history, threshold, and config.
     """
-    history_len = len(history_list)
-    if history_len == 0:
+    history_part_count = count_part_in_history_list(history_list)
+    if history_part_count == 0:
         return False
     enrichment_threshold = get_context_enrichment_threshold(
         ctx,
         context_enrichment_threshold_attr,
         render_context_enrichment_threshold,
     )
-    if enrichment_threshold == -1 or enrichment_threshold > history_len:
+    if enrichment_threshold == -1 or enrichment_threshold > history_part_count:
         return False
     return get_bool_attr(
         ctx,
@@ -158,9 +163,10 @@ async def maybe_enrich_context(
     context_enrichment_prompt: str,
 ) -> dict[str, Any]:
     """Enriches context based on history if enabled and threshold met."""
+    shorten_history_list = replace_system_prompt_in_history_list(history_list)
     if should_enrich_context(
         ctx,
-        history_list,
+        shorten_history_list,
         should_enrich_context_attr,
         render_enrich_context,
         context_enrichment_threshold_attr,
@@ -174,6 +180,6 @@ async def maybe_enrich_context(
                 prompt=context_enrichment_prompt,
             ),
             conversation_context=conversation_context,
-            history_list=history_list,
+            history_list=shorten_history_list,
         )
     return conversation_context

zrb/task/llm/history.py

@@ -1,6 +1,7 @@
 import json
 import os
 from collections.abc import Callable
+from copy import deepcopy
 from typing import Any, Optional
 
 from pydantic import BaseModel
@@ -177,66 +178,34 @@ async def write_conversation_history(
         write_file(history_file, history_data.model_dump_json(indent=2))
 
 
-async def prepare_initial_state(
-    ctx: AnyContext,
-    conversation_history_reader: (
-        Callable[[AnySharedContext], ConversationHistoryData | dict | list | None]
-        | None
-    ),
-    conversation_history_file_attr: StrAttr | None,
-    render_history_file: bool,
-    conversation_history_attr: (
-        ConversationHistoryData
-        | Callable[[AnySharedContext], ConversationHistoryData | dict | list]
-        | dict
-        | list
-    ),
-    conversation_context_getter: Callable[[AnyContext], dict[str, Any]],
-) -> tuple[ListOfDict, dict[str, Any]]:
-    """Reads history and prepares the initial conversation context."""
-    history_data: ConversationHistoryData = await read_conversation_history(
-        ctx,
-        conversation_history_reader,
-        conversation_history_file_attr,
-        render_history_file,
-        conversation_history_attr,
-    )
-    # Clean the history list to remove context from historical user prompts
-    cleaned_history_list = []
-    for interaction in history_data.history:
-        cleaned_history_list.append(
-            remove_context_from_interaction_history(interaction)
-        )
-    conversation_context = conversation_context_getter(ctx)
-    # Merge history context from loaded data without overwriting existing keys
-    for key, value in history_data.context.items():
-        if key not in conversation_context:
-            conversation_context[key] = value
-    # Return the CLEANED history list
-    return cleaned_history_list, conversation_context
-
-
-def remove_context_from_interaction_history(
-    interaction: dict[str, Any],
-) -> dict[str, Any]:
-    try:
-        cleaned_interaction = json.loads(json.dumps(interaction))
-    except Exception:
-        # Fallback to shallow copy if not JSON serializable (less safe)
-        cleaned_interaction = interaction.copy()
-    if "parts" in cleaned_interaction and isinstance(
-        cleaned_interaction["parts"], list
-    ):
-        for part in cleaned_interaction["parts"]:
-            is_user_prompt = part.get("part_kind") == "user-prompt"
-            has_str_content = isinstance(part.get("content"), str)
-            if is_user_prompt and has_str_content:
-                content = part["content"]
-                user_message_marker = "# User Message\n"
-                marker_index = content.find(user_message_marker)
-                if marker_index != -1:
-                    # Extract message after the marker and strip whitespace
-                    start_index = marker_index + len(user_message_marker)
-                    part["content"] = content[start_index:].strip()
-                # else: If marker not found, leave content as is (old format/error)
-    return cleaned_interaction
+def replace_system_prompt_in_history_list(
+    history_list: ListOfDict, replacement: str = "<main LLM system prompt>"
+) -> ListOfDict:
+    """
+    Returns a new history list where any part with part_kind 'system-prompt'
+    has its 'content' replaced with the given replacement string.
+    Args:
+        history: List of history items (each item is a dict with a 'parts' list).
+        replacement: The string to use in place of system-prompt content.
+
+    Returns:
+        A deep-copied list of history items with system-prompt content replaced.
+    """
+    new_history = deepcopy(history_list)
+    for item in new_history:
+        parts = item.get("parts", [])
+        for part in parts:
+            if part.get("part_kind") == "system-prompt":
+                part["content"] = replacement
+    return new_history
+
+
+def count_part_in_history_list(history_list: ListOfDict) -> int:
+    """Calculates the total number of 'parts' in a history list."""
+    history_part_len = 0
+    for history in history_list:
+        if "parts" in history:
+            history_part_len += len(history["parts"])
+        else:
+            history_part_len += 1
+    return history_part_len

zrb/task/llm/history_summarization.py

@@ -7,8 +7,13 @@ from zrb.attr.type import BoolAttr, IntAttr
 from zrb.context.any_context import AnyContext
 from zrb.llm_config import llm_config
 from zrb.task.llm.agent import run_agent_iteration
+from zrb.task.llm.history import (
+    count_part_in_history_list,
+    replace_system_prompt_in_history_list,
+)
 from zrb.task.llm.typing import ListOfDict
 from zrb.util.attr import get_bool_attr, get_int_attr
+from zrb.util.cli.style import stylize_faint
 
 if TYPE_CHECKING:
     from pydantic_ai.models import Model
@@ -18,17 +23,6 @@ else:
     ModelSettings = Any
 
 
-def get_history_part_len(history_list: ListOfDict) -> int:
-    """Calculates the total number of 'parts' in a history list."""
-    history_part_len = 0
-    for history in history_list:
-        if "parts" in history:
-            history_part_len += len(history["parts"])
-        else:
-            history_part_len += 1
-    return history_part_len
-
-
 def get_history_summarization_threshold(
     ctx: AnyContext,
     history_summarization_threshold_attr: IntAttr | None,
@@ -60,15 +54,15 @@ def should_summarize_history(
     render_history_summarization_threshold: bool,
 ) -> bool:
     """Determines if history summarization should occur based on length and config."""
-    history_part_len = get_history_part_len(history_list)
-    if history_part_len == 0:
+    history_part_count = count_part_in_history_list(history_list)
+    if history_part_count == 0:
         return False
     summarization_threshold = get_history_summarization_threshold(
         ctx,
         history_summarization_threshold_attr,
         render_history_summarization_threshold,
     )
-    if summarization_threshold == -1 or summarization_threshold > history_part_len:
+    if summarization_threshold == -1 or summarization_threshold > history_part_count:
         return False
     return get_bool_attr(
         ctx,
@@ -111,15 +105,18 @@ async def summarize_history(
     try:
         context_json = json.dumps(conversation_context)
         history_to_summarize_json = json.dumps(history_list)
-        summarization_user_prompt = (
-            f"# Current Context\n{context_json}\n\n"
-            f"# Conversation History to Summarize\n{history_to_summarize_json}"
+        summarization_user_prompt = "\n".join(
+            [
+                f"Current Context: {context_json}",
+                f"Conversation History to Summarize: {history_to_summarize_json}",
+            ]
         )
     except Exception as e:
         ctx.log_warning(f"Error formatting context/history for summarization: {e}")
         return conversation_context  # Return original context if formatting fails
 
     try:
+        ctx.print(stylize_faint("[Summarization Triggered]"), plain=True)
         summary_run = await run_agent_iteration(
             ctx=ctx,
             agent=summarization_agent,
@@ -128,6 +125,8 @@ async def summarize_history(
         )
         if summary_run and summary_run.result.output:
             summary_text = str(summary_run.result.output)
+            usage = summary_run.result.usage()
+            ctx.print(stylize_faint(f"[Token Usage] {usage}"), plain=True)
             # Update context with the new summary
             conversation_context["history_summary"] = summary_text
             ctx.log_info("History summarized and added/updated in context.")
@@ -153,9 +152,10 @@ async def maybe_summarize_history(
     summarization_prompt: str,
 ) -> tuple[ListOfDict, dict[str, Any]]:
     """Summarizes history and updates context if enabled and threshold met."""
+    shorten_history_list = replace_system_prompt_in_history_list(history_list)
     if should_summarize_history(
         ctx,
-        history_list,
+        shorten_history_list,
         should_summarize_history_attr,
         render_summarize_history,
         history_summarization_threshold_attr,
@@ -170,7 +170,7 @@ async def maybe_summarize_history(
                 prompt=summarization_prompt,
             ),
             conversation_context=conversation_context,
-            history_list=history_list,  # Pass the full list for context
+            history_list=shorten_history_list,  # Pass the full list for context
         )
         # Truncate the history list after summarization
         return [], updated_context

zrb/task/llm/prompt.py

@@ -1,11 +1,6 @@
-import json
-from textwrap import dedent
-from typing import Any
-
 from zrb.attr.type import StrAttr
 from zrb.context.any_context import AnyContext
 from zrb.llm_config import llm_config as default_llm_config
-from zrb.task.llm.context import extract_default_context
 from zrb.util.attr import get_attr, get_str_attr
 
 
@@ -127,25 +122,3 @@ def get_context_enrichment_prompt(
     if context_enrichment_prompt is not None:
         return context_enrichment_prompt
     return default_llm_config.get_default_context_enrichment_prompt()
-
-
-def build_user_prompt(
-    ctx: AnyContext,
-    message_attr: StrAttr | None,
-    conversation_context: dict[str, Any],
-) -> str:
-    """Constructs the final user prompt including context."""
-    original_user_message = get_user_message(ctx, message_attr)
-    # Combine default context, conversation context (potentially enriched/summarized)
-    modified_user_message, default_context = extract_default_context(
-        original_user_message
-    )
-    enriched_context = {**default_context, **conversation_context}
-    return dedent(
-        f"""
-        # Context
-        {json.dumps(enriched_context)}
-        # User Message
-        {modified_user_message}
-        """
-    ).strip()

zrb/task/llm_task.py

@@ -22,26 +22,24 @@ from zrb.input.any_input import AnyInput
 from zrb.task.any_task import AnyTask
 from zrb.task.base_task import BaseTask
 from zrb.task.llm.agent import get_agent, run_agent_iteration
-
-# No longer need llm_config here
 from zrb.task.llm.config import (
     get_model,
     get_model_settings,
 )
-from zrb.task.llm.context import get_conversation_context
+from zrb.task.llm.context import extract_default_context, get_conversation_context
 from zrb.task.llm.context_enrichment import maybe_enrich_context
 from zrb.task.llm.history import (
     ConversationHistoryData,
     ListOfDict,
-    prepare_initial_state,
+    read_conversation_history,
     write_conversation_history,
 )
 from zrb.task.llm.history_summarization import maybe_summarize_history
 from zrb.task.llm.prompt import (
-    build_user_prompt,
     get_combined_system_prompt,
     get_context_enrichment_prompt,
     get_summarization_prompt,
+    get_user_message,
 )
 from zrb.util.cli.style import stylize_faint
 from zrb.xcom.xcom import Xcom
@@ -240,6 +238,7 @@ class LLMTask(BaseTask):
             summarization_prompt_attr=self._summarization_prompt,
             render_summarization_prompt=self._render_summarization_prompt,
         )
+        user_message = get_user_message(ctx, self._message)
         # Get the combined system prompt using the new getter
         system_prompt = get_combined_system_prompt(
             ctx=ctx,
@@ -250,17 +249,19 @@ class LLMTask(BaseTask):
             special_instruction_prompt_attr=self._special_instruction_prompt,
             render_special_instruction_prompt=self._render_special_instruction_prompt,
         )
-        # 1. Prepare initial state (read history, get initial context)
-        history_list, conversation_context = await prepare_initial_state(
+        # 1. Prepare initial state (read history from previous session)
+        conversation_history = await read_conversation_history(
             ctx=ctx,
             conversation_history_reader=self._conversation_history_reader,
             conversation_history_file_attr=self._conversation_history_file,
             render_history_file=self._render_history_file,
             conversation_history_attr=self._conversation_history,
-            conversation_context_getter=lambda c: get_conversation_context(
-                c, self._conversation_context
-            ),
         )
+        history_list = conversation_history.history
+        conversation_context = {
+            **conversation_history.context,
+            **get_conversation_context(ctx, self._conversation_context),
+        }
         # 2. Enrich context (optional)
         conversation_context = await maybe_enrich_context(
             ctx=ctx,
@@ -289,18 +290,21 @@ class LLMTask(BaseTask):
             model_settings=model_settings,
             summarization_prompt=summarization_prompt,
         )
-        # 4. Build the final user prompt
-        user_prompt = build_user_prompt(
-            ctx=ctx,
-            message_attr=self._message,
-            conversation_context=conversation_context,
+        # 4. Build the final user prompt and system prompt
+        final_user_prompt, default_context = extract_default_context(user_message)
+        final_system_prompt = "\n".join(
+            [
+                system_prompt,
+                "# Context",
+                json.dumps({**default_context, **conversation_context}),
+            ]
         )
         # 5. Get the agent instance
         agent = get_agent(
             ctx=ctx,
             agent_attr=self._agent,
             model=model,
-            system_prompt=system_prompt,
+            system_prompt=final_system_prompt,
             model_settings=model_settings,
             tools_attr=self._tools,
             additional_tools=self._additional_tools,
@@ -309,7 +313,7 @@ class LLMTask(BaseTask):
         )
         # 6. Run the agent iteration and save the results/history
         return await self._run_agent_and_save_history(
-            ctx, agent, user_prompt, history_list, conversation_context
+            ctx, agent, final_user_prompt, history_list, conversation_context
         )
 
     async def _run_agent_and_save_history(
@@ -346,7 +350,7 @@ class LLMTask(BaseTask):
                     ctx.xcom[xcom_usage_key] = Xcom([])
                 usage = agent_run.result.usage()
                 ctx.xcom.get(xcom_usage_key).push(usage)
-                ctx.print(stylize_faint(f"[USAGE] {usage}"), plain=True)
+                ctx.print(stylize_faint(f"[Token Usage] {usage}"), plain=True)
                 return agent_run.result.output
             else:
                 ctx.log_warning("Agent run did not produce a result.")

{zrb-1.6.0.dist-info → zrb-1.6.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: zrb
-Version: 1.6.0
+Version: 1.6.1
 Summary: Your Automation Powerhouse
 Home-page: https://github.com/state-alchemists/zrb
 License: AGPL-3.0-or-later

{zrb-1.6.0.dist-info → zrb-1.6.1.dist-info}/RECORD

@@ -245,7 +245,7 @@ zrb/input/option_input.py,sha256=TQB82ko5odgzkULEizBZi0e9TIHEbIgvdP0AR3RhA74,213
 zrb/input/password_input.py,sha256=szBojWxSP9QJecgsgA87OIYwQrY2AQ3USIKdDZY6snU,1465
 zrb/input/str_input.py,sha256=NevZHX9rf1g8eMatPyy-kUX3DglrVAQpzvVpKAzf7bA,81
 zrb/input/text_input.py,sha256=NdceGgtbFFZkAoWOBW4xarjM-N1oePTStBJb8RbOqLw,3499
-zrb/llm_config.py,sha256=sFK4KCMSEDR4HLvutDLGdvG078llXvxop4F1qViE_Co,9269
+zrb/llm_config.py,sha256=TAGha1PAiS-vvxM3UyAlIK_0mh3qe5clOfRKbQhGUF8,9477
 zrb/runner/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 zrb/runner/cli.py,sha256=Fh1p5qwa8sK2IlHbPoBHWLBA7rplDMlQpZfmbMq6gy8,6835
 zrb/runner/common_util.py,sha256=JDMcwvQ8cxnv9kQrAoKVLA40Q1omfv-u5_d5MvvwHeE,1373
@@ -317,18 +317,18 @@ zrb/task/base_trigger.py,sha256=WSGcmBcGAZw8EzUXfmCjqJQkz8GEmi1RzogpF6A1V4s,6902
 zrb/task/cmd_task.py,sha256=3JFkWZEhyrQAwbQJs2pgICBmkohUR9T-hjXw82JyNtA,10720
 zrb/task/http_check.py,sha256=Gf5rOB2Se2EdizuN9rp65HpGmfZkGc-clIAlHmPVehs,2565
 zrb/task/llm/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-zrb/task/llm/agent.py,sha256=6DxmhezQ6Y_LqXJ24DqfABVPjv_Spo3Vvz5TH5M3IoQ,5470
+zrb/task/llm/agent.py,sha256=6wGSsw03GdY_fj12CsJh7wxB6BnE13N8RYXaWfbiUsk,5451
 zrb/task/llm/config.py,sha256=R6mkbm4d5ecN4KjjZaXbqNq9-8bXfdUGl_BML8hUWqY,3205
-zrb/task/llm/context.py,sha256=s313jXooOp8ht7j2sc3-d8xyW7oWs9bFv4KEyjicwys,3893
-zrb/task/llm/context_enrichment.py,sha256=lLPCtSSdB0jyReB3rk5taNvFPpXMmjDZ_Ro_Fz5GX68,5963
+zrb/task/llm/context.py,sha256=U9a8lxa2ikz6my0Sd5vpO763legHrMHyvBjbrqNmv0Y,3838
+zrb/task/llm/context_enrichment.py,sha256=dyUYhgozHZBQo-pDSoBkrlt39hOxS0i2IzDUKpPVF5U,6458
 zrb/task/llm/error.py,sha256=27DQXSG8SH1-XuvXFdZQKzP39wZDWmd_YnSTz6DJKKI,3690
-zrb/task/llm/history.py,sha256=LnrJdXLyo2qz-bNCwLorhoqGmgSiPTUU0bzY63w67-E,9257
-zrb/task/llm/history_summarization.py,sha256=0cWChp4OE_OiaNDhHRWi4rwHKTHsqWLYy3pS5IAIHpQ,6293
+zrb/task/llm/history.py,sha256=3WMXoi7RquxosXQf3iv2_BCeF8iKtY1f407pR71xERs,7745
+zrb/task/llm/history_summarization.py,sha256=UTyw7JF6cDmOJhuQOYVumF65QIqJHc3eQOa8p9Buwgw,6444
 zrb/task/llm/print_node.py,sha256=bpISOUxSH_JBLR-4Nq6-iLrzNWFagrKFX6u8ogYYMw8,4395
-zrb/task/llm/prompt.py,sha256=0uN96mPiq5o28Bz_f7wWmOD6-DJYD-nc_N5RovYz3Js,4719
+zrb/task/llm/prompt.py,sha256=56qabihh9ZaDF-QJkOW2EUJ4_hhLsRZxzkZ8nAHrrz0,3911
 zrb/task/llm/tool_wrapper.py,sha256=Xygd4VCY3ykjVv63pqlTI16ZG41ySkp683_5VTnL-Zo,6481
 zrb/task/llm/typing.py,sha256=c8VAuPBw_4A3DxfYdydkgedaP-LU61W9_wj3m3CAX1E,58
-zrb/task/llm_task.py,sha256=9r8SitsqBEOYGF92h_E8ROhofKXOh-BzrDIDyfg4-S8,15232
+zrb/task/llm_task.py,sha256=Bj_EJRv2qrCw6bgt7aHbaZwS5u0kc1R6NlqLkRw-0SE,15516
 zrb/task/make_task.py,sha256=PD3b_aYazthS8LHeJsLAhwKDEgdurQZpymJDKeN60u0,2265
 zrb/task/rsync_task.py,sha256=GSL9144bmp6F0EckT6m-2a1xG25AzrrWYzH4k3SVUKM,6370
 zrb/task/scaffolder.py,sha256=rME18w1HJUHXgi9eTYXx_T2G4JdqDYzBoNOkdOOo5-o,6806
@@ -369,7 +369,7 @@ zrb/util/string/name.py,sha256=SXEfxJ1-tDOzHqmSV8kvepRVyMqs2XdV_vyoh_9XUu0,1584
 zrb/util/todo.py,sha256=VGISej2KQZERpornK-8X7bysp4JydMrMUTnG8B0-liI,20708
 zrb/xcom/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 zrb/xcom/xcom.py,sha256=o79rxR9wphnShrcIushA0Qt71d_p3ZTxjNf7x9hJB78,1571
-zrb-1.6.0.dist-info/METADATA,sha256=7Py3F2SacChvaWLjNETTyNNqcP0OtKpDDO-sXW9j20E,8385
-zrb-1.6.0.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
-zrb-1.6.0.dist-info/entry_points.txt,sha256=-Pg3ElWPfnaSM-XvXqCxEAa-wfVI6BEgcs386s8C8v8,46
-zrb-1.6.0.dist-info/RECORD,,
+zrb-1.6.1.dist-info/METADATA,sha256=Mvuy3HzhIukGZ6QTM3X9RbFhQz6UFXhdngsC53dqSYQ,8385
+zrb-1.6.1.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+zrb-1.6.1.dist-info/entry_points.txt,sha256=-Pg3ElWPfnaSM-XvXqCxEAa-wfVI6BEgcs386s8C8v8,46
+zrb-1.6.1.dist-info/RECORD,,