repo stringclasses 454
values | file_path stringlengths 5 201 | extension stringclasses 1
value | content stringlengths 8 509k | num_lines int64 3 16.9k | size_bytes int64 8 511k |
|---|---|---|---|---|---|
QwenPaw | src/qwenpaw/agents/tools/_lsp_client.py | .py | # -*- coding: utf-8 -*-
"""Minimal LSP JSON-RPC client over subprocess stdio.
Designed for short, synchronous queries from inside the agent's
``lsp`` tool: definition, references, hover, document / workspace
symbols, implementation. No streaming, no incremental sync, no
diagnostics — only correct request / response p... | 481 | 14,851 |
QwenPaw | src/qwenpaw/agents/tools/_lsp_servers.py | .py | # -*- coding: utf-8 -*-
"""LSP server specs + discovery chain.
Each supported language has a :class:`LspServerSpec` whose ``discover``
callable returns the argv used to spawn the language server, or
``None`` if no server is reachable.
The Python entry has a guaranteed bundled fallback
(``python -m pylsp``) so it is a... | 160 | 4,966 |
QwenPaw | src/qwenpaw/agents/tools/make_skill_tools.py | .py | # -*- coding: utf-8 -*-
"""Tool backing the ``/make-skill`` flow."""
from __future__ import annotations
import json
import logging
import re
from agentscope.message import TextBlock
from agentscope.tool import ToolChunk
from agentscope.message import ToolResultState
from ...config.context import get_current_workspa... | 305 | 11,329 |
QwenPaw | src/qwenpaw/agents/tools/browser.py | .py | # -*- coding: utf-8 -*-
"""QwenPaw builtin tool binding for the unified Browser SDK."""
from __future__ import annotations
from pathlib import Path
from agentscope.tool import ToolChunk
from ...runtime.tool_registry import tool_descriptor
from ...browser.tool_entrypoint import (
_BROWSER_TOOL_DESCRIPTION,
... | 35 | 949 |
QwenPaw | src/qwenpaw/agents/tools/get_token_usage.py | .py | # -*- coding: utf-8 -*-
"""Tool to query token usage statistics."""
from datetime import date, timedelta
from agentscope.message import TextBlock
from agentscope.tool import ToolChunk
from agentscope.message import ToolResultState
from ...runtime.tool_registry import tool_descriptor
from ...token_usage import get_to... | 97 | 3,108 |
QwenPaw | src/qwenpaw/agents/tools/shell.py | .py | # -*- coding: utf-8 -*-
# flake8: noqa: E501
# pylint: disable=line-too-long
"""The shell command tool."""
import asyncio
import locale
import os
import re
import signal
import subprocess
import sys
import tempfile
import threading
import time
from dataclasses import dataclass, replace
from pathlib import Path
from ty... | 1,222 | 42,061 |
QwenPaw | src/qwenpaw/agents/tools/run_tool_batch.py | .py | # -*- coding: utf-8 -*-
"""Run a batch of tool calls with action-result references and control flow."""
from __future__ import annotations
import asyncio
import ast
import json
import logging
import operator
import re
import uuid
from collections.abc import Callable
from pathlib import Path
from typing import Any
fr... | 1,260 | 40,996 |
QwenPaw | src/qwenpaw/agents/tools/deprecated_browser/browser_snapshot.py | .py | # -*- coding: utf-8 -*-
"""Build role snapshot + refs for the deprecated browser tool."""
import re
from typing import Any
INTERACTIVE_ROLES = frozenset(
{
"button",
"link",
"textbox",
"checkbox",
"radio",
"combobox",
"listbox",
"menuitem",
"... | 249 | 6,876 |
QwenPaw | src/qwenpaw/agents/tools/deprecated_browser/__init__.py | .py | # -*- coding: utf-8 -*-
"""Deprecated structured-browser escape hatch.
This package preserves the stable ``browser`` implementation while the
experimental unified browser gathers beta feedback. It must not import the
unified browser SDK; the tools package chooses exactly one implementation.
"""
| 8 | 298 |
QwenPaw | src/qwenpaw/agents/tools/deprecated_browser/browser_control.py | .py | # -*- coding: utf-8 -*-
# flake8: noqa: E501
# pylint: disable=redefined-outer-name
"""Deprecated browser automation tool using Playwright.
Single tool with action-based API matching browser MCP: start, stop, open,
navigate, navigate_back, screenshot, snapshot, click, type, eval, evaluate,
resize, console_messages, ha... | 5,630 | 188,056 |
QwenPaw | src/qwenpaw/agents/context/types.py | .py | # -*- coding: utf-8 -*-
"""Shared data types for the context-management package."""
from __future__ import annotations
from dataclasses import dataclass, field
from typing import Any, Literal
class ContextWindowUnfitError(RuntimeError):
"""Raised when context compaction cannot fit the model input window."""
... | 54 | 2,042 |
QwenPaw | src/qwenpaw/agents/context/__init__.py | .py | # -*- coding: utf-8 -*-
"""Pluggable context-management strategies.
The default agent behavior is AgentScope-native compression; injecting a
:class:`ContextManager` replaces it. The only built-in alternative today is the
*scroll* strategy (durable ``history.db`` + an in-context eviction index + a
sandboxed ``recall_hi... | 241 | 9,668 |
QwenPaw | src/qwenpaw/agents/context/base.py | .py | # -*- coding: utf-8 -*-
"""The pluggable context-manager interface.
A ``ContextManager`` is an injectable strategy that owns an agent's context
management. :class:`~qwenpaw.agents.react_agent.QwenPawAgent` delegates its
AgentScope hooks and provider-overflow recovery to it:
* ``_save_to_context`` -> :meth:`on_save` (... | 52 | 1,917 |
QwenPaw | src/qwenpaw/agents/context/visual_compression/__init__.py | .py | # -*- coding: utf-8 -*-
"""Request-time visual context compression package.
The engineering implementation is based in part on pxpipe:
https://github.com/teamchong/pxpipe
"""
| 8 | 177 |
QwenPaw | src/qwenpaw/agents/context/visual_compression/config.py | .py | # -*- coding: utf-8 -*-
"""Code-owned Visual Compact policy and compression presets."""
from __future__ import annotations
from dataclasses import dataclass
from types import MappingProxyType
from typing import Literal, Mapping, cast
VisualCompressionEffort = Literal["low", "medium", "high"]
CANVAS_WIDTH = 1568
CAN... | 120 | 3,184 |
QwenPaw | src/qwenpaw/agents/context/visual_compression/pipeline/request.py | .py | # -*- coding: utf-8 -*-
"""Provider-independent context planning and request transformation.
The transformer works on deep copies of ``Msg`` objects. Images remain
AgentScope ``DataBlock`` instances until the active formatter builds the wire.
Tool documentation follows one provider-independent QwenPaw-native policy.
"... | 142 | 4,718 |
QwenPaw | src/qwenpaw/agents/context/visual_compression/pipeline/receipt.py | .py | # -*- coding: utf-8 -*-
"""Transform result and request-local exact recovery sources."""
from __future__ import annotations
import hashlib
from dataclasses import dataclass, field
from typing import Any
@dataclass
class CompressionReceipt:
"""Exact sources and aggregate observability for accepted replacements."... | 71 | 1,919 |
QwenPaw | src/qwenpaw/agents/context/visual_compression/pipeline/tool_schemas.py | .py | # -*- coding: utf-8 -*-
"""QwenPaw-native tool documentation compression.
QwenPaw's built-in tools are ordinary Python functions. AgentScope derives
their input schemas from signatures, type hints, defaults, and docstrings.
The production policy below follows that contract instead of trying to be a
general-purpose JSO... | 104 | 3,624 |
QwenPaw | src/qwenpaw/agents/context/visual_compression/pipeline/__init__.py | .py | # -*- coding: utf-8 -*-
"""Pure request planning and transformation stages."""
| 3 | 79 |
QwenPaw | src/qwenpaw/agents/context/visual_compression/pipeline/static_context.py | .py | # -*- coding: utf-8 -*-
"""QwenPaw system/tool planning and atomic visual replacement."""
from __future__ import annotations
import json
import re
from typing import Any
from agentscope.message import Msg, TextBlock
from ..config import (
CHARS_PER_TEXT_TOKEN_FALLBACK,
EffortPreset,
)
from ..rendering impor... | 273 | 9,161 |
QwenPaw | src/qwenpaw/agents/context/visual_compression/pipeline/tool_results.py | .py | # -*- coding: utf-8 -*-
"""Successful tool-result selection, paging, and visual replacement."""
from __future__ import annotations
import math
import re
from typing import Any
from agentscope.message import Msg, TextBlock, ToolResultBlock, ToolResultState
from ..config import (
CANVAS_PADDING,
CANVAS_WIDTH,... | 501 | 16,458 |
QwenPaw | src/qwenpaw/agents/context/visual_compression/pipeline/messages.py | .py | # -*- coding: utf-8 -*-
"""AgentScope message serialization and text normalization."""
from __future__ import annotations
import base64
import json
import re
from dataclasses import dataclass
from typing import TYPE_CHECKING, Any, Iterable
from agentscope.message import (
Base64Source,
DataBlock,
Msg,
... | 244 | 7,536 |
QwenPaw | src/qwenpaw/agents/context/visual_compression/pipeline/history.py | .py | # -*- coding: utf-8 -*-
"""Append-only, protocol-closed fixed-grid history planning."""
from __future__ import annotations
import re
from dataclasses import dataclass
from typing import Any
from agentscope.message import (
DataBlock,
Msg,
TextBlock,
ThinkingBlock,
ToolCallBlock,
ToolResultBlo... | 447 | 14,245 |
QwenPaw | src/qwenpaw/agents/context/visual_compression/pipeline/precision.py | .py | # -*- coding: utf-8 -*-
"""Deterministic exact-value precision lane for rendered text."""
from __future__ import annotations
import math
import re
from dataclasses import dataclass
from ..config import (
FACTSHEET_MAX_CHUNK_CHARS,
FACTSHEET_MAX_DISTINCT,
FACTSHEET_MAX_ENTRIES,
FACTSHEET_MAX_SCAN_CHAR... | 237 | 7,284 |
QwenPaw | src/qwenpaw/agents/context/visual_compression/pipeline/budget.py | .py | # -*- coding: utf-8 -*-
"""Token counting and visual-versus-text request budget policy."""
from __future__ import annotations
import math
from dataclasses import dataclass
from typing import TYPE_CHECKING
from ..config import (
CANVAS_MAX_HEIGHT,
CANVAS_PADDING,
CHARS_PER_TEXT_TOKEN_FALLBACK,
IMAGE_C... | 160 | 4,680 |
QwenPaw | src/qwenpaw/agents/context/visual_compression/rendering/renderer.py | .py | # -*- coding: utf-8 -*-
"""Deterministic text layout and PNG rendering."""
from __future__ import annotations
import base64
import hashlib
import re
import struct
import sys
import zlib
from array import array
from dataclasses import dataclass, field, replace
from functools import lru_cache
from pathlib import Path
f... | 821 | 24,613 |
QwenPaw | src/qwenpaw/agents/context/visual_compression/rendering/__init__.py | .py | # -*- coding: utf-8 -*-
"""Production surface for deterministic visual-context rendering."""
from .renderer import (
RenderedPage,
estimate_text_pages,
measure_content_columns,
page_count_for_text,
prepare_render_text,
render_cache_info,
render_rows_per_page,
render_text_pages,
)
__all... | 25 | 539 |
QwenPaw | src/qwenpaw/agents/context/visual_compression/runtime/middleware.py | .py | # -*- coding: utf-8 -*-
"""AgentScope pre-model hook for request-time visual compression."""
from __future__ import annotations
import asyncio
import logging
import time
from concurrent.futures import ThreadPoolExecutor
from functools import partial
from typing import TYPE_CHECKING, Any, Callable
from agentscope.mid... | 236 | 7,949 |
QwenPaw | src/qwenpaw/agents/context/visual_compression/runtime/__init__.py | .py | # -*- coding: utf-8 -*-
"""Framework adapters and request-local visual-compression state."""
| 3 | 93 |
QwenPaw | src/qwenpaw/agents/context/visual_compression/runtime/recovery.py | .py | # -*- coding: utf-8 -*-
"""Turn-local source storage and exact visual-context recovery tool."""
from __future__ import annotations
from collections.abc import Awaitable, Callable
import logging
import time
from typing import Any
from agentscope.message import TextBlock, ToolResultState
from agentscope.tool import To... | 518 | 17,617 |
QwenPaw | src/qwenpaw/agents/context/scroll/_as_internals.py | .py | # -*- coding: utf-8 -*-
"""Thin adapter over the AgentScope 2.0 *private* methods the scroll manager
depends on.
AgentScope 2.0 is alpha; ``_prepare_model_input`` and
``_split_context_for_compression`` are underscore-prefixed internals that may
be renamed or have their signatures changed between releases. Routing ever... | 36 | 1,334 |
QwenPaw | src/qwenpaw/agents/context/scroll/eviction_index.py | .py | # -*- coding: utf-8 -*-
"""The eviction index — an in-context, tier-capped odometer of evicted turns.
The whole index lives in the prompt as ONE placeholder, so the model always
*sees the map* of what it evicted. The structure is a stack of tiers:
Tier 0 (bottom) the newest evictions; each block lists its turns ... | 422 | 16,324 |
QwenPaw | src/qwenpaw/agents/context/scroll/manager.py | .py | # -*- coding: utf-8 -*-
"""ScrollContextManager — write-through + eviction-index context management.
The strategy form of the design in ``CONTEXT_MANAGEMENT.html``: instead of
subclassing the agent, it is injected into :class:`QwenPawAgent` and drives the
two delegated hooks.
* :meth:`on_save` — every live turn is pe... | 2,026 | 82,336 |
QwenPaw | src/qwenpaw/agents/context/scroll/__init__.py | .py | # -*- coding: utf-8 -*-
"""Scroll context strategy: durable history + eviction index + recall REPL."""
| 3 | 103 |
QwenPaw | src/qwenpaw/agents/context/scroll/memoryspace.py | .py | # -*- coding: utf-8 -*-
"""The model's SQLite working surface inside ``recall_history_python``.
Self-contained (stdlib only) so the sandboxed REPL cell can import it by bare
module name, without the rest of qwenpaw on the path.
``main`` is an in-memory database the model owns read/write — its scratch
space. The durab... | 2,016 | 77,149 |
QwenPaw | src/qwenpaw/agents/context/scroll/sync.py | .py | # -*- coding: utf-8 -*-
"""Import raw ``sessions/*.json`` conversation history into ``history.db``.
Conversations are saved one-file-per-session under ``{workspace}/sessions/``.
Under the ``scroll`` strategy, durable recall lives in ``history.db``,
populated live by :class:`ScrollContextManager`. Messages from before ... | 908 | 31,792 |
QwenPaw | src/qwenpaw/agents/context/scroll/repl.py | .py | # -*- coding: utf-8 -*-
"""The sandboxed ``recall_history_python`` recall tool.
This is raw conversation-history recall — the agent's own recorded turns
across its sessions (verbatim). The model recalls history by running Python
here, not by scrolling back. Each call runs a fresh process (Option A:
stateless cells) in... | 304 | 12,699 |
QwenPaw | src/qwenpaw/agents/context/scroll/continuation_summary.py | .py | # -*- coding: utf-8 -*-
"""Plain-text continuation summaries with code-managed durable provenance.
The model is deliberately asked for ordinary Markdown, never structured
output or JSON. Parsing into this internal representation happens locally and
fails closed, so provider/model formatting quirks cannot break compact... | 756 | 30,482 |
QwenPaw | src/qwenpaw/agents/context/scroll/history.py | .py | # -*- coding: utf-8 -*-
"""Durable, file-backed conversation history shared across sessions."""
from __future__ import annotations
import json
import logging
import sqlite3
import sys
import threading
from collections.abc import Sequence
from datetime import datetime, timezone
from pathlib import Path
from typing imp... | 761 | 30,804 |
QwenPaw | src/qwenpaw/agents/context/scroll/serialize.py | .py | # -*- coding: utf-8 -*-
"""Serialize AgentScope ``Msg`` blocks into ``conversation_history`` rows."""
from __future__ import annotations
import re
from copy import deepcopy
from dataclasses import dataclass
from typing import Any
from agentscope.message import Msg
from ....constant import QWENPAW_MESSAGE_TAG_KEY
fro... | 348 | 12,981 |
QwenPaw | src/qwenpaw/agents/context/scroll/prompt.py | .py | # -*- coding: utf-8 -*-
"""System-prompt block taught to the agent under the scroll strategy.
Injected only when ``strategy == "scroll"`` (see
:class:`qwenpaw.runtime.prompt_contributors.ScrollContextContributor`). It
teaches what the model must know for the eviction index to be useful: how to
write useful milestone h... | 204 | 13,060 |
QwenPaw | src/qwenpaw/agents/context/scroll/recall_tool.py | .py | # -*- coding: utf-8 -*-
"""The structured ``recall_history`` recall tool.
The parameterized front door to durable history: ``expand`` / ``search`` /
``recall_tool`` cover the overwhelming share of recall calls, and none of
them needs model-authored code — each is a bound, read-only SQL query over
:class:`.memoryspace.... | 1,061 | 38,425 |
QwenPaw | src/qwenpaw/agents/utils/media_token_estimate.py | .py | # -*- coding: utf-8 -*-
"""Context-ring token estimates for inlined media payloads.
Local heuristic only — not a billing formula. Never treats base64 bytes
as text tokens (``len(data) // 4``).
"""
from __future__ import annotations
import base64
import hashlib
import io
import math
import re
import wave
EMPTY_MEDIA... | 184 | 4,955 |
QwenPaw | src/qwenpaw/agents/utils/as_msg_handler.py | .py | # -*- coding: utf-8 -*-
# pylint: disable=too-many-branches,too-many-statements
"""Token counting and block-level statistics for AgentScope messages.
Used by :func:`context_stats.format_history_str` to render the ``/history``
command output with per-message token breakdowns.
"""
import json
import logging
from agent... | 258 | 9,835 |
QwenPaw | src/qwenpaw/agents/utils/context_stats.py | .py | # -*- coding: utf-8 -*-
"""Token accounting helpers for ``/history`` and context inspection."""
from __future__ import annotations
from typing import TYPE_CHECKING
from .as_msg_handler import AsMsgHandler
from .estimate_token_counter import EstimatedTokenCounter
if TYPE_CHECKING:
from agentscope.state import Age... | 85 | 2,661 |
QwenPaw | src/qwenpaw/agents/utils/tool_message_utils.py | .py | # -*- coding: utf-8 -*-
"""Tool message validation and sanitization utilities.
This module ensures tool_call/tool_result messages are properly
paired and ordered to prevent API errors.
Supports both dict blocks (1.x ``type="tool_use"``) and Pydantic
``ToolCallBlock``/``ToolResultBlock`` objects (2.0 ``type="tool_call... | 648 | 22,599 |
QwenPaw | src/qwenpaw/agents/utils/setup_utils.py | .py | # -*- coding: utf-8 -*-
"""Setup and initialization utilities for agent configuration.
This module handles copying markdown configuration files to
the working directory.
"""
import logging
import shutil
from pathlib import Path
from ...constant import SUPPORTED_AGENT_LANGUAGES
logger = logging.getLogger(__name__)
_... | 273 | 8,065 |
QwenPaw | src/qwenpaw/agents/utils/__init__.py | .py | # -*- coding: utf-8 -*-
"""
Agent utilities package.
This package provides utilities for agent operations:
- audio_transcription: Audio file transcription
- estimate_token_counter: Estimated token counting
- file_handling: File download and management
- message_processing: Message content manipulation and validation
-... | 115 | 3,061 |
QwenPaw | src/qwenpaw/agents/utils/message_processing.py | .py | # -*- coding: utf-8 -*-
"""Message processing utilities for agent communication.
This module handles:
- File and media block processing
- Message content manipulation
- Message validation
"""
import asyncio
import logging
import os
import urllib.parse
from pathlib import Path
from typing import Optional
from agentsco... | 601 | 19,289 |
QwenPaw | src/qwenpaw/agents/utils/registry.py | .py | # -*- coding: utf-8 -*-
"""Generic registry for registering and retrieving implementations."""
from typing import Callable, Generic, Type, TypeVar
T = TypeVar("T")
class Registry(Generic[T]):
"""Generic registry for registering and retrieving implementations.
Example:
from qwenpaw.agents.memory.bas... | 48 | 1,427 |
QwenPaw | src/qwenpaw/agents/utils/estimate_token_counter.py | .py | # -*- coding: utf-8 -*-
"""Estimated token counter implementation."""
# Standalone class — qwenpaw's context manager only calls
# ``await counter.count(text)``, so duck typing is enough; no base
# class needed.
class EstimatedTokenCounter:
"""Token counter that estimates tokens using character-based calculation.
... | 41 | 1,419 |
QwenPaw | src/qwenpaw/agents/utils/file_handling.py | .py | # -*- coding: utf-8 -*-
"""File handling utilities for downloading and managing files.
This module provides utilities for:
- Downloading files from base64 encoded data
- Downloading files from URLs
- Managing download directories
- Reading text files with encoding fallback for cross-platform compatibility
"""
import o... | 402 | 13,363 |
QwenPaw | src/qwenpaw/agents/utils/as_msg_stat.py | .py | # -*- coding: utf-8 -*-
"""Schema definitions for AgentScope message statistics."""
from pydantic import BaseModel, Field
from ...constant import TRUNCATION_NOTICE_MARKER
_DEFAULT_MAX_BLOCK_TEXT_PREVIEW_LENGTH = 100
_DEFAULT_MAX_FORMATTER_TEXT_LENGTH = 1000
class AsBlockStat(BaseModel):
"""Statistics and metad... | 133 | 4,505 |
QwenPaw | src/qwenpaw/agents/utils/token_counter.py | .py | # -*- coding: utf-8 -*-
"""Token counting utilities."""
from typing import TYPE_CHECKING
from .estimate_token_counter import EstimatedTokenCounter
if TYPE_CHECKING:
from qwenpaw.config.config import AgentProfileConfig
def get_token_counter(
agent_config: "AgentProfileConfig",
) -> EstimatedTokenCounter:
... | 27 | 714 |
QwenPaw | src/qwenpaw/agents/utils/message_request_normalizer.py | .py | # -*- coding: utf-8 -*-
"""Normalization helpers for provider chat payloads.
The persisted session history remains AgentScope ``Msg`` objects. For
provider requests we build a normalized copy before formatting so
request-time repair and multimodal downgrade logic does not mutate the
stored conversation state.
"""
fro... | 273 | 10,031 |
QwenPaw | src/qwenpaw/agents/utils/audio_transcription.py | .py | # -*- coding: utf-8 -*-
"""Audio transcription utility.
Transcribes audio files to text using either:
- An OpenAI-compatible ``/v1/audio/transcriptions`` endpoint (Whisper API), or
- The locally installed ``openai-whisper`` Python library (Local Whisper).
Transcription is only attempted when explicitly enabled via th... | 318 | 9,675 |
QwenPaw | src/qwenpaw/agents/utils/image_freezing.py | .py | # -*- coding: utf-8 -*-
"""Freeze local image references into immutable message content."""
from __future__ import annotations
import base64
from io import BytesIO
from pathlib import Path
from typing import Any
from urllib.parse import unquote, urlparse
from urllib.request import url2pathname
from agentscope.messag... | 225 | 7,325 |
QwenPaw | src/qwenpaw/agents/skills/__init__.py | .py | # -*- coding: utf-8 -*-
"""Agent skills directory."""
| 3 | 54 |
QwenPaw | src/qwenpaw/agents/skills/pptx-zh/scripts/thumbnail.py | .py | """Create thumbnail grids from PowerPoint presentation slides.
Creates a grid layout of slide thumbnails for quick visual analysis.
Labels each thumbnail with its XML filename (e.g., slide1.xml).
Hidden slides are shown with a placeholder pattern.
Usage:
python thumbnail.py input.pptx [output_prefix] [--cols N]
... | 307 | 9,630 |
QwenPaw | src/qwenpaw/agents/skills/pptx-zh/scripts/clean.py | .py | """Remove unreferenced files from an unpacked PPTX directory.
Usage: python clean.py <unpacked_dir>
Example:
python clean.py unpacked/
This script removes:
- Orphaned slides (not in sldIdLst) and their relationships
- [trash] directory (unreferenced files)
- Orphaned .rels files for deleted resources
- Unreferen... | 287 | 9,583 |
QwenPaw | src/qwenpaw/agents/skills/pptx-zh/scripts/add_slide.py | .py | """Add a new slide to an unpacked PPTX directory.
Usage: python add_slide.py <unpacked_dir> <source>
The source can be:
- A slide file (e.g., slide2.xml) - duplicates the slide
- A layout file (e.g., slideLayout2.xml) - creates from layout
Examples:
python add_slide.py unpacked/ slide2.xml
# Duplicates s... | 196 | 6,872 |
QwenPaw | src/qwenpaw/agents/skills/pptx-zh/scripts/office/unpack.py | .py | """Unpack Office files (DOCX, PPTX, XLSX) for editing.
Extracts the ZIP archive, pretty-prints XML files, and optionally:
- Merges adjacent runs with identical formatting (DOCX only)
- Simplifies adjacent tracked changes from same author (DOCX only)
Usage:
python unpack.py <office_file> <output_dir> [options]
Ex... | 133 | 4,052 |
QwenPaw | src/qwenpaw/agents/skills/pptx-zh/scripts/office/validate.py | .py | """
Command line tool to validate Office document XML files against XSD schemas and tracked changes.
Usage:
python validate.py <path> [--original <original_file>] [--auto-repair] [--author NAME]
The first argument can be either:
- An unpacked directory containing the Office document XML files
- A packed Office fi... | 112 | 3,668 |
QwenPaw | src/qwenpaw/agents/skills/pptx-zh/scripts/office/soffice.py | .py | """
Helper for running LibreOffice (soffice) in environments where AF_UNIX
sockets may be blocked (e.g., sandboxed VMs). Detects the restriction
at runtime and applies an LD_PRELOAD shim if needed.
Usage:
from office.soffice import run_soffice, get_soffice_cmd, get_soffice_env
# Option 1 – run soffice direct... | 219 | 6,607 |
QwenPaw | src/qwenpaw/agents/skills/pptx-zh/scripts/office/pack.py | .py | """Pack a directory into a DOCX, PPTX, or XLSX file.
Validates with auto-repair, condenses XML formatting, and creates the Office file.
Usage:
python pack.py <input_directory> <output_file> [--original <file>] [--validate true|false]
Examples:
python pack.py unpacked/ output.docx --original input.docx
py... | 160 | 4,991 |
QwenPaw | src/qwenpaw/agents/skills/pptx-zh/scripts/office/validators/redlining.py | .py | """
Validator for tracked changes in Word documents.
"""
import subprocess
import tempfile
import zipfile
from pathlib import Path
class RedliningValidator:
def __init__(self, unpacked_dir, original_docx, verbose=False, author="Claude"):
self.unpacked_dir = Path(unpacked_dir)
self.original_docx ... | 248 | 8,918 |
QwenPaw | src/qwenpaw/agents/skills/pptx-zh/scripts/office/validators/pptx.py | .py | """
Validator for PowerPoint presentation XML files against XSD schemas.
"""
import re
from .base import BaseSchemaValidator
class PPTXSchemaValidator(BaseSchemaValidator):
PRESENTATIONML_NAMESPACE = (
"http://schemas.openxmlformats.org/presentationml/2006/main"
)
ELEMENT_RELATIONSHIP_TYPES = ... | 276 | 9,824 |
QwenPaw | src/qwenpaw/agents/skills/pptx-zh/scripts/office/validators/docx.py | .py | """
Validator for Word document XML files against XSD schemas.
"""
import os
import random
import re
import tempfile
import zipfile
import defusedxml.minidom
import lxml.etree
from .base import BaseSchemaValidator
class DOCXSchemaValidator(BaseSchemaValidator):
WORD_2006_NAMESPACE = "http://schemas.openxmlfor... | 448 | 16,401 |
QwenPaw | src/qwenpaw/agents/skills/pptx-zh/scripts/office/validators/base.py | .py | """
Base validator with common validation logic for document files.
"""
import re
from pathlib import Path
import defusedxml.minidom
import lxml.etree
class BaseSchemaValidator:
IGNORED_VALIDATION_ERRORS = [
"hyphenationZone",
"purl.org/dc/terms",
]
UNIQUE_ID_REQUIREMENTS = {
"... | 848 | 32,651 |
QwenPaw | src/qwenpaw/agents/skills/pptx-zh/scripts/office/helpers/merge_runs.py | .py | """Merge adjacent runs with identical formatting in DOCX.
Merges adjacent <w:r> elements that have identical <w:rPr> properties.
Works on runs in paragraphs and inside tracked changes (<w:ins>, <w:del>).
Also:
- Removes rsid attributes from runs (revision metadata that doesn't affect rendering)
- Removes proofErr ele... | 200 | 5,567 |
QwenPaw | src/qwenpaw/agents/skills/pptx-zh/scripts/office/helpers/simplify_redlines.py | .py | """Simplify tracked changes by merging adjacent w:ins or w:del elements.
Merges adjacent <w:ins> elements from the same author into a single element.
Same for <w:del> elements. This makes heavily-redlined documents easier to
work with by reducing the number of tracked change wrappers.
Rules:
- Only merges w:ins with ... | 198 | 5,754 |
QwenPaw | src/qwenpaw/agents/skills/docx-en/scripts/comment.py | .py | """Add comments to DOCX documents.
Usage:
python comment.py unpacked/ 0 "Comment text"
python comment.py unpacked/ 1 "Reply text" --parent 0
Text should be pre-escaped XML (e.g., & for &, ’ for smart quotes).
After running, add markers to document.xml:
<w:commentRangeStart w:id="0"/>
... comme... | 319 | 10,694 |
QwenPaw | src/qwenpaw/agents/skills/docx-en/scripts/accept_changes.py | .py | """Accept all tracked changes in a DOCX file using LibreOffice.
Requires LibreOffice (soffice) to be installed.
"""
import argparse
import logging
import shutil
import subprocess
import tempfile
from pathlib import Path
from office.soffice import get_soffice_cmd, get_soffice_env
logger = logging.getLogger(__name__)... | 139 | 4,228 |
QwenPaw | src/qwenpaw/agents/skills/docx-en/scripts/office/soffice.py | .py | """
Helper for running LibreOffice (soffice) in environments where AF_UNIX
sockets may be blocked (e.g., sandboxed VMs). Detects the restriction
at runtime and applies an LD_PRELOAD shim if needed.
Usage:
from office.soffice import run_soffice, get_soffice_cmd, get_soffice_env
# Option 1 – run soffice direct... | 221 | 6,772 |
QwenPaw | src/qwenpaw/agents/skills/xlsx-en/scripts/recalc.py | .py | """
Excel Formula Recalculation Script
Recalculates all formulas in an Excel file using LibreOffice
"""
import json
import os
import platform
import subprocess
import sys
from pathlib import Path
from office.soffice import get_soffice_cmd, get_soffice_env
from openpyxl import load_workbook
MACRO_DIR_MACOS = "~/Libr... | 210 | 6,526 |
QwenPaw | src/qwenpaw/agents/memory/agent_md_manager.py | .py | # -*- coding: utf-8 -*-
"""Agent Markdown manager for reading and writing markdown files in working
and memory directories."""
from datetime import datetime, timezone
from pathlib import Path
from typing import Literal
from ..utils.file_handling import read_text_file_with_encoding_fallback
from ...config.config impor... | 322 | 11,857 |
QwenPaw | src/qwenpaw/agents/memory/__init__.py | .py | # -*- coding: utf-8 -*-
"""Memory management module for QwenPaw agents."""
from typing import TYPE_CHECKING
from .agent_md_manager import AgentMdManager
from .base_memory_manager import BaseMemoryManager
from .reme_light_memory_manager import ReMeLightMemoryManager
from .adbpg_memory_manager import (
ADBPGMemoryM... | 73 | 2,120 |
QwenPaw | src/qwenpaw/agents/memory/reme_light_memory_manager.py | .py | # -*- coding: utf-8 -*-
"""ReMe-backed memory manager for agents.
The public class and registry key keep the historical ``ReMeLight`` naming so
existing agent configs continue to work, but the implementation delegates to
ReMe's application/job framework.
"""
import asyncio
import hashlib
import logging
import os
from... | 1,256 | 46,349 |
QwenPaw | src/qwenpaw/agents/memory/reme_config.py | .py | # -*- coding: utf-8 -*-
"""Embedded ReMe application configuration for QwenPaw memory.
ReMe's standalone CLI normally loads YAML such as
``reme/config/default.yaml`` or ``reme/config/qwenpaw.yaml``. QwenPaw embeds
ReMe as an in-process application, so it passes an equivalent configuration
dict directly to ``reme.appl... | 724 | 27,777 |
QwenPaw | src/qwenpaw/agents/memory/embedding_model.py | .py | # -*- coding: utf-8 -*-
"""AgentScope embedding model construction and connectivity checks."""
from __future__ import annotations
import asyncio
import math
import time
from dataclasses import dataclass
from typing import Any
from agentscope.credential import (
DashScopeCredential,
GeminiCredential,
Olla... | 156 | 4,917 |
QwenPaw | src/qwenpaw/agents/memory/dummy.py | .py | # -*- coding: utf-8 -*-
"""No-op memory manager – disables all memory functionality."""
from collections.abc import Callable
from agentscope.middleware import MiddlewareBase
from agentscope.tool import ToolChunk
from .base_memory_manager import BaseMemoryManager, memory_registry
@memory_registry.register("none")
cl... | 43 | 1,293 |
QwenPaw | src/qwenpaw/agents/memory/adbpg_client.py | .py | # -*- coding: utf-8 -*-
"""REST client for ADBPG memory storage."""
import json
import logging
from dataclasses import dataclass
import httpx
logger = logging.getLogger(__name__)
@dataclass
class ADBPGConfig:
"""ADBPG REST API configuration."""
rest_base_url: str
rest_api_key: str
search_timeout: ... | 157 | 4,809 |
QwenPaw | src/qwenpaw/agents/memory/adbpg_prompts.py | .py | # -*- coding: utf-8 -*-
# flake8: noqa: E501
"""ADBPG memory guidance prompts."""
ADBPG_MEMORY_GUIDANCE_ZH = """\
## 记忆
每次会话都是全新的。你的长期记忆存储在云端数据库（ADBPG）中，对话中的重要信息会自动提取并保存。
### 🧠 云端长期记忆
- 你的对话内容会**自动保存**到云端长期记忆中
- 系统会自动从对话中提取事实、偏好、决策等关键信息
- 无需手动写入文件，记忆提取由服务端 LLM 完成
- 记忆跨会话持久化 — 即使会话结束，记忆仍然保留
### 🔍 检索记忆
回答关于过往工作、决... | 71 | 3,275 |
QwenPaw | src/qwenpaw/agents/memory/adbpg_memory_manager.py | .py | # -*- coding: utf-8 -*-
"""ADBPG Memory Manager for QwenPaw agents.
Provides long-term memory backed by AnalyticDB for PostgreSQL (ADBPG).
Context compaction is handled natively by AgentScope's
``Agent.compress_context()``; tool result pruning is handled by
``ToolResultPruningMiddleware``. This class only manages long... | 431 | 14,720 |
QwenPaw | src/qwenpaw/agents/memory/base_memory_manager.py | .py | # -*- coding: utf-8 -*-
"""Abstract base class for memory managers."""
import asyncio
import json
import logging
import uuid
from abc import ABC, abstractmethod
from collections.abc import Callable
from copy import deepcopy
from datetime import datetime, timezone
from typing import Any
from agentscope.message import ... | 695 | 24,398 |
QwenPaw | src/qwenpaw/agents/memory/prompts.py | .py | # -*- coding: utf-8 -*-
"""Memory guidance prompts."""
MEMORY_GUIDANCE = {
"zh": (
"# 长期记忆\n\n"
"工作目录下保存着你的个人知识库。当问题涉及用户过去的事实、偏好、"
"决策或经验时，先使用 `memory_search` 检索相关信息。检索结果中会包含"
"相关内容片段及其文件路径；如果片段不足以回答问题，再使用 `read_file` "
"按路径渐进式展开，只读取当前任务所需的内容。"
),
"en": (
"# ... | 34 | 1,375 |
QwenPaw | src/qwenpaw/agents/memory/proactive/proactive_prompts.py | .py | # -*- coding: utf-8 -*-
"""Prompt templates for proactive conversation feature."""
PROACTIVE_TASK_EXTRACTION_PROMPT = (
"You are given the user's recent session contexts and current screen "
"content (if provided) in a reversed order.\n"
"\n"
"Your job:\n"
"1. Find 1–3 likely high-level goals the u... | 74 | 2,989 |
QwenPaw | src/qwenpaw/agents/memory/proactive/proactive_types.py | .py | # -*- coding: utf-8 -*-
"""Type definitions for proactive conversation feature."""
from dataclasses import dataclass
from typing import Optional
from datetime import datetime
@dataclass
class ProactiveConfig:
"""Configuration for proactive feature."""
enabled: bool = False
idle_minutes: int = 30
las... | 38 | 841 |
QwenPaw | src/qwenpaw/agents/memory/proactive/proactive_trigger.py | .py | # -*- coding: utf-8 -*-
"""Trigger logic for proactive conversation feature."""
import asyncio
import logging
from datetime import datetime, timezone
from typing import TYPE_CHECKING, Dict, Optional, Any
from .proactive_types import ProactiveConfig
from .proactive_responder import generate_proactive_response
from .pr... | 338 | 10,592 |
QwenPaw | src/qwenpaw/agents/memory/proactive/__init__.py | .py | # -*- coding: utf-8 -*-
"""Proactive memory submodule for CoPaw agents."""
from .proactive_types import (
ProactiveConfig,
ProactiveTask,
ProactiveQueryResult,
)
from .proactive_trigger import (
enable_proactive_for_session,
disable_proactive_for_session,
proactive_trigger_loop,
proactive_t... | 31 | 753 |
QwenPaw | src/qwenpaw/agents/memory/proactive/proactive_responder.py | .py | # -*- coding: utf-8 -*-
"""Responder logic for proactive conversation feature."""
import asyncio
import logging
import re
from datetime import datetime, timezone
from typing import TYPE_CHECKING, Optional, List, Dict
import aiohttp
from agentscope.agent import Agent, ReActConfig
from agentscope.message import Msg, T... | 389 | 12,035 |
QwenPaw | src/qwenpaw/agents/memory/proactive/proactive_utils.py | .py | # -*- coding: utf-8 -*-
"""Utility functions for proactive messaging features."""
from copy import deepcopy
import json
import logging
from datetime import datetime, timedelta, timezone
from typing import TYPE_CHECKING, List, Optional, Any
from agentscope.agent import Agent
from agentscope.message import Msg, TextBlo... | 449 | 13,644 |
QwenPaw | src/qwenpaw/agents/skill_system/hub.py | .py | # -*- coding: utf-8 -*-
"""Skills hub client and install helpers."""
from __future__ import annotations
import asyncio
import base64
import contextvars
import io
import json
import logging
import os
import re
import time
import zipfile
from collections import OrderedDict
from contextlib import asynccontextmanager, con... | 2,344 | 72,938 |
QwenPaw | src/qwenpaw/agents/skill_system/models.py | .py | # -*- coding: utf-8 -*-
"""Shared models and constants for the skill system."""
from __future__ import annotations
from dataclasses import dataclass
from pathlib import Path
from typing import Any
from pydantic import BaseModel, Field
from ...exceptions import SkillConflictError
ALL_SKILL_ROUTING_CHANNELS = [
... | 81 | 1,760 |
QwenPaw | src/qwenpaw/agents/skill_system/store.py | .py | # -*- coding: utf-8 -*-
"""Local skill storage, manifest, metadata, and filesystem helpers."""
from __future__ import annotations
import hashlib
import io
import json
import logging
import os
import re
import shutil
import tempfile
import time
import zipfile
from collections.abc import Callable, Iterator
from context... | 1,141 | 35,846 |
QwenPaw | src/qwenpaw/agents/skill_system/__init__.py | .py | # -*- coding: utf-8 -*-
"""Skill system exports."""
from .models import (
SkillConflictError,
SkillInfo,
)
from .pool_service import SkillPoolService, run_pool_auto_update_sync
from .registry import (
apply_skill_config_env_overrides,
ensure_skill_pool_initialized,
ensure_skills_initialized,
re... | 48 | 1,220 |
QwenPaw | src/qwenpaw/agents/skill_system/registry.py | .py | # -*- coding: utf-8 -*-
"""Skill registries, builtin sync, runtime resolution, and env overrides."""
from __future__ import annotations
import hashlib
import json
import logging
import os
import re
import threading
from collections.abc import Iterator
from contextlib import contextmanager
from pathlib import Path
fro... | 1,525 | 51,368 |
QwenPaw | src/qwenpaw/agents/skill_system/pool_service.py | .py | # -*- coding: utf-8 -*-
"""Shared skill-pool lifecycle service."""
from __future__ import annotations
import hashlib
import logging
import shutil
from pathlib import Path
from typing import Any
from ...exceptions import SkillsError
from ..utils.file_handling import read_text_file_with_encoding_fallback
from .models ... | 1,264 | 43,749 |
QwenPaw | src/qwenpaw/agents/skill_system/workspace_service.py | .py | # -*- coding: utf-8 -*-
"""Workspace-scoped skill lifecycle service."""
from __future__ import annotations
import shutil
from pathlib import Path
from typing import Any
from ...exceptions import SkillsError
from ..utils.file_handling import read_text_file_with_encoding_fallback
from .models import SkillInfo
from .re... | 782 | 26,733 |
QwenPaw | src/qwenpaw/app/driver_config_service.py | .py | # -*- coding: utf-8 -*-
"""Application service for persisted Driver configuration."""
from __future__ import annotations
import asyncio
import logging
from pathlib import Path
from typing import Any
from fastapi import HTTPException
from ..drivers.capabilities import DriverRuntimeInfo
from ..drivers.contracts impor... | 227 | 7,592 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.