Skip to main content

An academic paper writing agent based on LangGraph

Project description

seele-scholar-agent

基于 LangGraph 的学术论文写作智能体。自动生成结构化论文大纲、撰写章节、支持人工审核与多轮修订。

功能特性

  • 研究检索:从 ArXiv、Semantic Scholar、OpenAlex 搜索相关论文,支持注入自定义检索源(PubMed、IEEE Xplore、用户文献库等)
  • 选题推荐:基于文献趋势推荐具体可行的论文选题
  • 大纲规划:基于检索结果生成结构化论文大纲,并为每个章节标注建议图表
  • 章节撰写:结合 RAG 上下文自动撰写论文章节,在正文中插入图表占位符(含 chunk_id 绑定)
  • 审核修订:AI 审核机制,支持多轮修改
  • 一致性检查:检查各章节之间的术语、引用、逻辑一致性
  • 参考文献生成:自动调用 CrossRef API 验证 DOI、补全发表年份与期刊/会议信息,生成准确的标准格式参考文献列表;API 不可用时自动回退到本地提取
  • 流式调用:所有节点支持 astream() 方法,可实时输出 token
  • 多模型支持:支持 OpenAI、DeepSeek、Groq 及任何 OpenAI 兼容 API

安装

# 克隆仓库
git clone https://github.com/your-org/seele-scholar-agent.git
cd seele-scholar-agent

# 使用 uv 安装(推荐)
uv sync

# 或使用 pip 安装
pip install -e .

配置

Agent 包内部配置

seele-scholar-agent 只管理自身运行所需的极少数配置,通过 src/seele_scholar_agent/.env 加载:

cp src/seele_scholar_agent/.env.example src/seele_scholar_agent/.env
变量 说明 默认值
SEMANTIC_SCHOLAR_API_KEY Semantic Scholar API 密钥(可选,提升频率限制)
MAX_REVISIONS 最大修订轮次 3

调用方配置(由你的项目管理)

LLM、向量数据库等配置由调用方自行管理,在初始化时注入到 agent:

# 你的项目 .env(示例)
OPENAI_API_KEY=sk-...
OPENAI_MODEL=gpt-4o
OPENAI_BASE_URL=https://api.openai.com/v1

支持任何 OpenAI 兼容 API,通过 ChatOpenAI 构造参数传入:

DeepSeek:

OPENAI_API_KEY = "sk-..."
OPENAI_MODEL = "deepseek-chat"
OPENAI_BASE_URL = "https://api.deepseek.com/v1"

Groq(免费 Llama):

OPENAI_API_KEY = "gsk_..."
OPENAI_MODEL = "llama-3.1-70b-versatile"
OPENAI_BASE_URL = "https://api.groq.com/openai/v1"

使用方法

快速开始(完整工作流)

import asyncio
from datetime import datetime
from uuid import uuid4

from langchain_openai import ChatOpenAI
from seele_scholar_agent.config import settings
from seele_scholar_agent.graph import create_writing_graph
from seele_scholar_agent.state import AgentState
from seele_scholar_agent.agent_config import PromptsConfig


async def main():
    model = ChatOpenAI(
        model="gpt-4o",
        api_key="sk-...",
        base_url="https://api.openai.com/v1",
        temperature=0.7,
    )

    prompts = PromptsConfig(
        planner_system_prompt="你是大纲规划师...",
        planner_user_prompt="研究主题:{topic}\n...",
        writer_system_prompt="你是学术论文撰写者...",
        writer_user_prompt="论文主题:{topic}\n...",
        reviewer_system_prompt="你是审稿人...",
        reviewer_user_prompt="请审阅以下章节:{content}\n...",
        topic_proposer_system_prompt="你是学术导师...",
        topic_proposer_user_prompt="宽泛研究方向:{topic}\n...",
        finalizer_system_prompt="你是摘要撰写者...",
        finalizer_user_prompt="论文主题:{topic}\n...",
        consistency_check_system_prompt="你负责一致性检查...",
        consistency_check_user_prompt="请检查章节一致性:{sections_summary}\n...",
        citation_alignment_system_prompt="你负责引用核查...",
        citation_alignment_user_prompt="请检查引用:{content}\n...",
        topic_translation_system_prompt="You are an academic search query expert...",
        topic_translation_user_prompt="Translate the following topic: {topic}",
    )

    app = create_writing_graph(model=model, prompts=prompts, rag_retriever=None)

    initial_state: AgentState = {
        "thread_id": str(uuid4()),
        "topic": "你的研究主题",
        "language": "zh",
        "created_at": datetime.now(),
        "tenant_id": None,
        "broad_papers": [],
        "proposed_topics": [],
        "papers": [],
        "search_queries": [],
        "outline": None,
        "outline_approved": False,
        "sections": [],
        "current_section_index": 0,
        "sections_completed": [],
        "review_history": [],
        "current_review": None,
        "rag_context": [],
        "status": "idle",
        "error_message": None,
        "max_revisions": settings.MAX_REVISIONS,
        "revision_count": 0,
        "references": [],
        "consistency_issues": [],
        "consistency_checked": False,
    }

    result = await app.ainvoke(
        initial_state,
        config={"configurable": {"thread_id": initial_state["thread_id"]}}
    )

    print(f"状态: {result.get('status')}")
    if result.get("outline"):
        print(f"标题: {result['outline'].title}")
        for s in result["outline"].sections:
            print(f"  {s.order}. {s.title}")
            for fig in s.suggested_figures:
                print(f"    [图表] {fig}")


if __name__ == "__main__":
    asyncio.run(main())

结合 RAG 使用(WriterNode 文献注入)

RAG 上下文通过 RAGRetrieverFunc 回调注入,由调用方实现检索逻辑(Qdrant、Chroma 等均可):

from seele_scholar_agent import RAGRetrieverFunc
from seele_scholar_agent.state import DocumentChunk

# RAGRetrieverFunc = Callable[[str], Awaitable[list[DocumentChunk]]]
# 调用方实现,根据查询返回文档块列表

async def my_rag_retriever(query: str) -> list[DocumentChunk]:
    # 使用 Qdrant、Chroma 等向量数据库检索
    results = await qdrant_store.asimilarity_search(query, k=5)
    return [
        DocumentChunk(
            chunk_id=doc.metadata["chunk_id"],
            content=doc.page_content,
            source=doc.metadata.get("source", ""),
        )
        for doc in results
    ]

app = create_writing_graph(
    model=model,
    prompts=prompts,
    rag_retriever=my_rag_retriever,
)

注入自定义论文检索源(PaperSearchFunc)

通过 extra_paper_retrievers 参数可以注入任意外部论文检索源(PubMed、IEEE Xplore、用户私有文献库等),结果与内置三源(ArXiv、Semantic Scholar、OpenAlex)自动合并去重、按相关度排序:

from seele_scholar_agent import PaperSearchFunc
from seele_scholar_agent.state import PaperMetadata

# PaperSearchFunc = Callable[[str], Awaitable[list[PaperMetadata]]]

async def pubmed_retriever(query: str) -> list[PaperMetadata]:
    """从 PubMed 检索论文"""
    results = await fetch_pubmed(query)
    return [
        PaperMetadata(
            paper_id=f"pubmed:{r['pmid']}",
            title=r["title"],
            authors=r["authors"],
            abstract=r["abstract"],
            url=f"https://pubmed.ncbi.nlm.nih.gov/{r['pmid']}/",
            relevance_score=0.0,   # ResearcherNode 会重新计算排序
            source="user_library",  # 建议使用 "user_library" 标记外部来源
        )
        for r in results
    ]

async def my_library_retriever(query: str) -> list[PaperMetadata]:
    """从私有文献库检索"""
    ...

app = create_writing_graph(
    model=model,
    prompts=prompts,
    rag_retriever=my_rag_retriever,
    extra_paper_retrievers=[pubmed_retriever, my_library_retriever],
)

断点与恢复执行

create_writing_graphtopic_proposerplanner 节点后设置了断点,分别等待用户选择选题和确认大纲。

人工确认模式:

thread_id = str(uuid4())
config = {"configurable": {"thread_id": thread_id}}

# 第一次调用,运行到 topic_proposer 断点
result = await app.ainvoke(initial_state, config=config)

# 用户选择选题
if result["status"] == "waiting_human":
    for t in result["proposed_topics"]:
        print(f"- {t.title}{t.difficulty_level})")
    chosen_topic = result["proposed_topics"][0].title

    # 更新选题并继续到 planner 断点
    app.update_state(config, {"topic": chosen_topic})
    result = await app.ainvoke(None, config=config)

# 用户确认大纲
if result["status"] == "waiting_human":
    outline = result["outline"]
    print(f"标题: {outline.title}")
    for s in outline.sections:
        print(f"  {s.order}. {s.title} — 建议图表: {s.suggested_figures}")

    app.update_state(config, {"outline_approved": True})
    result = await app.ainvoke(None, config=config)

自动模式(跳过断点,适合测试):

# 使用 create_simple_writing_graph,无断点
from seele_scholar_agent.graph import create_simple_writing_graph

app = create_simple_writing_graph(model=model, prompts=prompts, rag_retriever=None)
result = await app.ainvoke(initial_state, config=config)

单节点流式调用

所有节点均支持独立实例化和流式调用,适用于只需要某个功能的场景。

流式事件结构(NodeStreamEvent)

# NodeStreamEvent 是 TypedDict(total=False),type 字段区分事件类型:
# - "token"    : LLM 输出的文本片段,携带 token: str
# - "progress" : 阶段进度提示,携带 progress: str
# - "result"   : 最终结果,携带 result: dict,始终是最后一个事件

async for event in node.astream(state):
    if event["type"] == "token":
        print(event["token"], end="", flush=True)
    elif event["type"] == "progress":
        print(f"\n[{event['progress']}]")
    elif event["type"] == "result":
        data = event.get("result", {})  # 注意:total=False,用 .get() 访问

论文检索(ResearcherNode)

from seele_scholar_agent import ResearcherNode

researcher = ResearcherNode(
    llm=model,
    extra_paper_retrievers=[pubmed_retriever],  # 可选
)

async for event in researcher.astream(state):
    if event["type"] == "progress":
        print(f"[{event['progress']}]")
    elif event["type"] == "result":
        papers = event.get("result", {}).get("papers", [])
        for p in papers:
            print(f"[{p.source}] {p.title}")

选题推荐(TopicProposerNode)

from seele_scholar_agent import TopicProposerNode

proposer = TopicProposerNode(llm=model, prompts=prompts)

async for event in proposer.astream(state):
    if event["type"] == "result":
        for topic in event.get("result", {}).get("proposed_topics", []):
            print(f"- {topic.title}{topic.difficulty_level})")

大纲规划(PlannerNode)

from seele_scholar_agent import PlannerNode

planner = PlannerNode(llm=model, prompts=prompts)

async for event in planner.astream(state):
    if event["type"] == "token":
        print(event["token"], end="", flush=True)
    elif event["type"] == "result":
        outline = event.get("result", {}).get("outline")
        for section in outline.sections:
            print(f"{section.order}. {section.title}")
            for fig in section.suggested_figures:
                print(f"   [建议图表] {fig}")

章节撰写(WriterNode)

from seele_scholar_agent import WriterNode

writer = WriterNode(llm=model, prompts=prompts, rag_retriever=my_rag_retriever)

async for event in writer.astream(state):
    if event["type"] == "token":
        print(event["token"], end="", flush=True)
    elif event["type"] == "result":
        sections = event.get("result", {}).get("sections", [])
        print(sections[state["current_section_index"]].content)

审稿(ReviewerNode)

from seele_scholar_agent import ReviewerNode

reviewer = ReviewerNode(llm=model, prompts=prompts)

async for event in reviewer.astream(state):
    if event["type"] == "result":
        review = event.get("result", {}).get("current_review")
        print(f"通过: {review.approved}, 评分: {review.score}")
        for issue in review.issues:
            print(f"  [{issue.type}] {issue.description}")

图表占位符处理

Writer 节点在正文中插入图表占位符,格式如下:

{{FIGURE: 条形图,展示各模型在ImageNet上的Top-1准确率对比 | chunks:[abc123,def456]}}
{{TABLE: 各方法时间复杂度与空间复杂度对比 | chunks:[xyz789]}}
  • FIGURE / TABLE:图表类型
  • 描述部分:图表内容和展示目的
  • chunks:对应 RAG 中数据来源的 chunk_id 列表,由 LLM 在写作时从 RAG context 中自动绑定;无数据支撑时为空数组 chunks:[]

主项目解析示例:

import re

FIGURE_PATTERN = re.compile(
    r'\{\{(FIGURE|TABLE): (.+?) \| chunks:\[([^\]]*)\]\}\}'
)

def extract_figures(content: str):
    results = []
    for fig_type, description, chunks_str in FIGURE_PATTERN.findall(content):
        chunk_ids = [c.strip() for c in chunks_str.split(",") if c.strip()]
        results.append({
            "type": fig_type,
            "description": description,
            "chunk_ids": chunk_ids,
        })
    return results

async def render_figures(content: str, rag_store) -> str:
    for fig_type, description, chunks_str in FIGURE_PATTERN.findall(content):
        chunk_ids = [c.strip() for c in chunks_str.split(",") if c.strip()]
        data = await rag_store.get_chunks(chunk_ids)
        placeholder = f"{{{{{fig_type}: {description} | chunks:[{chunks_str}]}}}}"
        content = content.replace(placeholder, render_chart(fig_type, description, data))
    return content

节点返回值(result 事件字段)

每个节点 astream() 的最后一个事件类型为 "result"event.get("result", {}) 是一个 dict,对应 AgentState 的局部更新。

TopicProposerNode

字段 类型 说明
broad_papers list[PaperMetadata] 宽泛检索阶段找到的文献
proposed_topics list[ProposedTopic] 推荐的论文选题列表
status "waiting_human" 等待用户选择选题

ProposedTopic 字段:

字段 类型 说明
title str 选题标题
description str 详细描述和切入点
trend_analysis str 趋势分析(受哪些文献启发)
difficulty_level "easy" | "medium" | "hard" 难度评估

ResearcherNode

字段 类型 说明
papers list[PaperMetadata] 检索到的论文列表(多源去重,按相关度排序)
search_queries list[str] 实际使用的搜索词列表
status "planning" 进入规划阶段

PaperMetadata 字段:

字段 类型 说明
paper_id str 论文唯一 ID
title str 论文标题
authors list[str] 作者列表
abstract str 摘要
url str | None 论文页面链接
pdf_url str | None PDF 直链
relevance_score float 相关度分数
source "arxiv" | "semantic_scholar" | "openalex" | "user_library" 来源

PlannerNode

字段 类型 说明
outline OutlineStructure 生成的论文大纲
sections list[SectionDraft] 按大纲拆分的章节草稿(初始均为 pending)
current_section_index 0 重置章节索引
status "waiting_human" 等待用户确认大纲

SectionOutline 字段:

字段 类型 说明
title str 章节标题
description str 章节描述
order int 章节顺序编号
key_points list[str] 关键论点列表
suggested_figures list[str] 建议在本章插入的图表描述列表

WriterNode

字段 类型 说明
sections list[SectionDraft] 当前章节 content 已填充,status 变为 "review"
status "reviewing" 进入审稿阶段

SectionDraft 字段:

字段 类型 说明
section_id str 章节 ID(如 "section_0"
title str 章节标题
content str 正文内容(Markdown,含图表占位符)
order_index int 章节顺序
status "pending" | "writing" | "review" | "approved" | "auto_generated" 章节状态
revision_count int 已修订次数
review_comments list[str] 审稿意见(修订时注入)

ReviewerNode

审核通过时:

字段 类型 说明
sections list[SectionDraft] 当前章节 status 更新为 "approved"
sections_completed list[str] 新增当前章节标题
review_history list[dict] 新增本次审核记录
current_review dict 本次审核结果(见下表)
current_section_index int 推进到下一章节
status "writing" | "completed" 继续或完成

审核不通过时:

字段 类型 说明
sections list[SectionDraft] 当前章节 status 变回 "writing",追加审稿意见
revision_count int 修订计数加 1
status "writing" 返回撰写阶段重写

current_review dict 字段:

字段 类型 说明
approved bool 是否通过
score int 评分(1-10)
issues list[dict] 问题列表(含 type、description、suggestion、location)
summary str 总体审阅意见

issues[].type 可选值:factual_error | missing_citation | weak_argument | format_issue | citation_mismatch | other

FinalizerNode

字段 类型 说明
sections list[SectionDraft](可选) 更新了摘要/结论内容(status 为 "auto_generated"
status "completed" 工作流完成

ConsistencyCheckerNode

字段 类型 说明
consistency_checked True 标记检查已完成
consistency_issues list[ConsistencyIssue] 发现的一致性问题(无问题时为空数组)

ConsistencyIssue 字段:

字段 类型 说明
issue_type "terminology" | "citation" | "logic" | "other" 问题类型
description str 问题描述
sections_involved list[str] 涉及的章节标题列表
suggestion str 修改建议

ReferenceGeneratorNode

字段 类型 说明
references list[ReferenceEntry] 参考文献列表(按正文引用顺序排列)
status "completed" 工作流完成

ReferenceEntry 字段:

字段 类型 说明
number int 引用编号,对应正文中的 [N]
paper_id str 论文 ID
title str 论文标题
authors list[str] 作者列表
year int | None 发表年份
venue str | None 发表期刊/会议(由 CrossRef 补全)
url str | None 论文链接
doi str | None DOI 标识符
formatted str 格式化字符串(如 [1] Author. Title. Venue. (Year)

自定义 Prompts(PromptsConfig)

所有节点的 prompt 均通过 PromptsConfig 注入,支持完全自定义:

from seele_scholar_agent.agent_config import PromptsConfig

prompts = PromptsConfig(
    planner_system_prompt="你是大纲规划师...",
    planner_user_prompt="研究主题:{topic}\n...",
    writer_system_prompt="你是学术论文撰写者...",
    writer_user_prompt="论文主题:{topic}\n...",
    reviewer_system_prompt="你是审稿人...",
    reviewer_user_prompt="请审阅以下章节:{content}\n...",
    topic_proposer_system_prompt="你是学术导师...",
    topic_proposer_user_prompt="宽泛研究方向:{topic}\n...",
    finalizer_system_prompt="你是摘要撰写者...",
    finalizer_user_prompt="论文主题:{topic}\n...",
    consistency_check_system_prompt="你负责一致性检查...",
    consistency_check_user_prompt="请检查章节一致性:{sections_summary}\n...",
    citation_alignment_system_prompt="你负责引用核查...",
    citation_alignment_user_prompt="请检查引用:{content}\n...",
    topic_translation_system_prompt="You are an academic search query expert...",
    topic_translation_user_prompt="Translate the following topic: {topic}",
)

项目结构

src/seele_scholar_agent/
├── __init__.py             # 公共 API 导出
├── config.py               # 配置管理
├── state.py                # Pydantic 模型和 TypedDict 状态定义
├── graph.py                # LangGraph 工作流定义
├── agent_config.py         # PromptsConfig、RAGRetrieverFunc、PaperSearchFunc
├── logging.py              # 结构化日志配置
├── i18n.py                 # 多语言支持
├── tools/
│   └── crossref.py         # CrossRef REST API 查询(DOI 验证、元数据补全)
└── nodes/
    ├── __init__.py         # NodeStreamEvent、_stream_llm_text、invoke_with_retry
    ├── topic_proposer.py   # 选题推荐节点
    ├── researcher.py       # 论文检索节点(ArXiv、Semantic Scholar、OpenAlex + 自定义源)
    ├── planner.py          # 大纲规划节点(含 suggested_figures)
    ├── writer.py           # 章节撰写节点(含图表占位符插入)
    ├── reviewer.py         # 审稿节点
    ├── finalizer.py        # 摘要/结论生成节点
    ├── consistency_checker.py  # 一致性检查节点
    ├── reference_generator.py  # 参考文献生成节点(含 CrossRef 集成)
    └── (prompts 已移除,通过 PromptsConfig 由调用方注入)

AgentState 状态字段

字段 类型 说明
thread_id str 线程 ID,用于对话持久化
topic str 研究主题
language Literal["zh","en","ja"] 输出语言,默认 "zh"
created_at datetime 创建时间
tenant_id str | None 租户 ID(多租户场景)
broad_papers list[PaperMetadata] 选题阶段检索到的宽泛文献
proposed_topics list[ProposedTopic] 推荐的论文选题列表
papers list[PaperMetadata] 正式写作阶段检索到的论文
search_queries list[str] 搜索查询记录
outline OutlineStructure | None 生成的大纲(含各章节 suggested_figures
outline_approved bool 大纲是否已审核通过
sections list[SectionDraft] 拆分的章节列表
current_section_index int 当前正在撰写的章节索引
sections_completed list[str] 已完成的章节标题列表
review_history list[dict] 审核历史记录
current_review ReviewResult | None 当前审核结果
rag_context list[DocumentChunk] RAG 检索到的上下文
references list[ReferenceEntry] 生成的参考文献列表
consistency_issues list[ConsistencyIssue] 一致性检查发现的问题
consistency_checked bool 是否已完成一致性检查
status Literal[...] 当前工作流状态
error_message str | None 错误信息
max_revisions int 最大修订次数
revision_count int 当前修订计数

status 可选值: idle | researching | planning | writing | reviewing | finalizing | checking_consistency | waiting_human | completed | failed

工作流程

START → topic_proposer → [选题确认] → researcher → planner → [大纲确认] → writer → reviewer
                                                                                      ↓
                                                                                 [审核通过?]
                                                                                      ↓
                                                                           writer(下一节) 或 finalizer
                                                                                      ↓
                                                                           consistency_checker → reference_generator → END
  1. TopicProposer:基于宽泛研究方向推荐 3 个具体选题(断点,等待用户选择)
  2. Researcher:从 ArXiv、Semantic Scholar、OpenAlex 及自定义检索源检索相关论文
  3. Planner:生成结构化大纲,每章节附 suggested_figures(断点,等待用户确认)
  4. Writer:根据大纲和 RAG 上下文撰写章节,正文中插入图表占位符(绑定 chunk_id)
  5. Reviewer:审核章节质量,支持多轮修订
  6. Finalizer:生成摘要和结论
  7. ConsistencyChecker:检查全文一致性
  8. ReferenceGenerator:生成参考文献列表

create_simple_writing_graph 构建无断点版工作流,适合全自动运行场景。

开发

# 安装开发依赖
uv sync --extra dev

# 运行代码检查
ruff check src/

# 运行类型检查
mypy src/

# 运行测试
pytest

许可证

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

seele_scholar_agent-0.9.0.tar.gz (228.8 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

seele_scholar_agent-0.9.0-py3-none-any.whl (35.9 kB view details)

Uploaded Python 3

File details

Details for the file seele_scholar_agent-0.9.0.tar.gz.

File metadata

  • Download URL: seele_scholar_agent-0.9.0.tar.gz
  • Upload date:
  • Size: 228.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.9.17 {"installer":{"name":"uv","version":"0.9.17","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for seele_scholar_agent-0.9.0.tar.gz
Algorithm Hash digest
SHA256 2bffd7227909d1ee12896c63494cf3ed03dbd30c663b528242caeac82cc39098
MD5 0ee278ce06834e7232d72c7bc462af0a
BLAKE2b-256 7784a8f6192b76cc9c6068b6522ae98c0c8bc9c99659a59771a037112b701bdf

See more details on using hashes here.

File details

Details for the file seele_scholar_agent-0.9.0-py3-none-any.whl.

File metadata

  • Download URL: seele_scholar_agent-0.9.0-py3-none-any.whl
  • Upload date:
  • Size: 35.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.9.17 {"installer":{"name":"uv","version":"0.9.17","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for seele_scholar_agent-0.9.0-py3-none-any.whl
Algorithm Hash digest
SHA256 7b3a96242ba3ff90cf307b9b93062b361bf8325e0e9c05b6de172e2ea473fbf3
MD5 0e942c5ac7594295b37101bc9a9ad80f
BLAKE2b-256 3ec7017fe6e47d878569ad6ed51514d2edbc5183360033b4277f01e7e06248f5

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page