Merge branch 'main' into reasoning

Author: better629

.gitignore

@@ -162,6 +162,8 @@ examples/graph_store.json
 examples/image__vector_store.json
 examples/index_store.json
 .chroma
+.chroma_exp_data
+.role_memory_data
 *~$*
 workspace/*
 tmp
@@ -189,6 +191,7 @@ cov.xml
 *-structure.json
 *.dot
 .python-version
+tests/data/requirements/*.jpg
 *.csv
 metagpt/ext/sela/results/*
 .chainlit/

Dockerfile

@@ -3,7 +3,7 @@ FROM nikolaik/python-nodejs:python3.9-nodejs20-slim
 
 # Install Debian software needed by MetaGPT and clean up in one RUN command to reduce image size
 RUN apt update &&\
-    apt install -y libgomp1 git chromium fonts-ipafont-gothic fonts-wqy-zenhei fonts-thai-tlwg fonts-kacst fonts-freefont-ttf libxss1 --no-install-recommends &&\
+    apt install -y libgomp1 git chromium fonts-ipafont-gothic fonts-wqy-zenhei fonts-thai-tlwg fonts-kacst fonts-freefont-ttf libxss1 --no-install-recommends file &&\
     apt clean && rm -rf /var/lib/apt/lists/*
 
 # Install Mermaid CLI globally

News.md

@@ -0,0 +1,22 @@
+## Earlier news
+
+🚀 Oct. 29, 2024: We introduced three papers: [AFLOW](https://arxiv.org/abs/2410.10762), [FACT](https://arxiv.org/abs/2410.21012), and [SELA](https://arxiv.org/abs/2410.17238), check the [code](examples)!
+
+🚀 Mar. 29, 2024: [v0.8.0](https://github.com/geekan/MetaGPT/releases/tag/v0.8.0) released. Now you can use Data Interpreter ([arxiv](https://arxiv.org/abs/2402.18679), [example](https://docs.deepwisdom.ai/main/en/DataInterpreter/), [code](https://github.com/geekan/MetaGPT/tree/main/examples/di)) via pypi package import. Meanwhile, we integrated the RAG module and supported multiple new LLMs.
+
+🚀 Feb. 08, 2024: [v0.7.0](https://github.com/geekan/MetaGPT/releases/tag/v0.7.0) released, supporting assigning different LLMs to different Roles. We also introduced [Data Interpreter](https://github.com/geekan/MetaGPT/blob/main/examples/di/README.md), a powerful agent capable of solving a wide range of real-world problems.
+
+🚀 Jan. 16, 2024: Our paper [MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework
+](https://openreview.net/forum?id=VtmBAGCN7o) accepted for **oral presentation (top 1.2%)** at ICLR 2024, **ranking #1** in the LLM-based Agent category.
+
+🚀 Jan. 03, 2024: [v0.6.0](https://github.com/geekan/MetaGPT/releases/tag/v0.6.0) released, new features include serialization, upgraded OpenAI package and supported multiple LLM, provided [minimal example for debate](https://github.com/geekan/MetaGPT/blob/main/examples/debate_simple.py) etc.
+
+🚀 Dec. 15, 2023: [v0.5.0](https://github.com/geekan/MetaGPT/releases/tag/v0.5.0) released, introducing some experimental features such as incremental development, multilingual, multiple programming languages, etc.
+
+🔥 Nov. 08, 2023: MetaGPT is selected into [Open100: Top 100 Open Source achievements](https://www.benchcouncil.org/evaluation/opencs/annual.html).
+
+🔥 Sep. 01, 2023: MetaGPT tops GitHub Trending Monthly for the **17th time** in August 2023.
+
+🌟 Jun. 30, 2023: MetaGPT is now open source.
+
+🌟 Apr. 24, 2023: First line of MetaGPT code committed.

README.md

@@ -27,30 +27,16 @@
 </p>
 
 ## News
-🚀 Feb. 19, 2025: Today we are officially launching our natural language programming product: MGX (MetaGPT X) - the world's first AI agent development team.  [Offical website](https://mgx.dev/) [Twitter](https://x.com/MetaGPT_/status/1892199535130329356)
 
-🚀 Jan. 22, 2025: Our paper [AFlow: Automating Agentic Workflow Generation](https://openreview.net/forum?id=z5uVAKwmjf) accepted for **oral presentation (top 1.8%)** at ICLR 2025, **ranking #2** in the LLM-based Agent category.
-
-🚀 Oct. 29, 2024: We introduced three papers: [AFLOW](https://arxiv.org/abs/2410.10762), [FACT](https://arxiv.org/abs/2410.21012), and [SELA](https://arxiv.org/abs/2410.17238), check the [code](examples)!
-
-🚀 Mar. 29, 2024: [v0.8.0](https://github.com/geekan/MetaGPT/releases/tag/v0.8.0) released. Now you can use Data Interpreter ([arxiv](https://arxiv.org/abs/2402.18679), [example](https://docs.deepwisdom.ai/main/en/DataInterpreter/), [code](https://github.com/geekan/MetaGPT/tree/main/examples/di)) via pypi package import. Meanwhile, we integrated the RAG module and supported multiple new LLMs.
-
-🚀 Feb. 08, 2024: [v0.7.0](https://github.com/geekan/MetaGPT/releases/tag/v0.7.0) released, supporting assigning different LLMs to different Roles. We also introduced [Data Interpreter](https://github.com/geekan/MetaGPT/blob/main/examples/di/README.md), a powerful agent capable of solving a wide range of real-world problems.
-
-🚀 Jan. 16, 2024: Our paper [MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework
-](https://openreview.net/forum?id=VtmBAGCN7o) accepted for **oral presentation (top 1.2%)** at ICLR 2024, **ranking #1** in the LLM-based Agent category.
+🚀 Mar. 4, 2025: 🎉 [mgx.dev](https://mgx.dev/) is the #1 Product of the Day on @ProductHunt! 🏆
 
-🚀 Jan. 03, 2024: [v0.6.0](https://github.com/geekan/MetaGPT/releases/tag/v0.6.0) released, new features include serialization, upgraded OpenAI package and supported multiple LLM, provided [minimal example for debate](https://github.com/geekan/MetaGPT/blob/main/examples/debate_simple.py) etc.
+🚀 Feb. 19, 2025: Today we are officially launching our natural language programming product: [MGX (MetaGPT X)](https://mgx.dev/) - the world's first AI agent development team. More details on [Twitter](https://x.com/MetaGPT_/status/1892199535130329356).
 
-🚀 Dec. 15, 2023: [v0.5.0](https://github.com/geekan/MetaGPT/releases/tag/v0.5.0) released, introducing some experimental features such as incremental development, multilingual, multiple programming languages, etc.
+🚀 Feb. 17, 2025: We introduced two papers: [SPO](https://arxiv.org/pdf/2502.06855) and [AOT](https://arxiv.org/pdf/2502.12018), check the [code](examples)!
 
-🔥 Nov. 08, 2023: MetaGPT is selected into [Open100: Top 100 Open Source achievements](https://www.benchcouncil.org/evaluation/opencs/annual.html).
-
-🔥 Sep. 01, 2023: MetaGPT tops GitHub Trending Monthly for the **17th time** in August 2023.
-
-🌟 Jun. 30, 2023: MetaGPT is now open source.
+🚀 Jan. 22, 2025: Our paper [AFlow: Automating Agentic Workflow Generation](https://openreview.net/forum?id=z5uVAKwmjf) accepted for **oral presentation (top 1.8%)** at ICLR 2025, **ranking #2** in the LLM-based Agent category.
 
-🌟 Apr. 24, 2023: First line of MetaGPT code committed.
+👉👉 [Earlier news](News.md) 
 
 ## Software Company as Multi-Agent System
 
@@ -75,6 +61,8 @@ pip install --upgrade metagpt
 # or `git clone https://github.com/geekan/MetaGPT && cd MetaGPT && pip install --upgrade -e .`
 ```
 
+**Install [node](https://nodejs.org/en/download) and [pnpm](https://pnpm.io/installation#using-npm) before actual use.**
+
 For detailed installation guidance, please refer to [cli_install](https://docs.deepwisdom.ai/main/en/guide/get_started/installation.html#install-stable-version)
  or [docker_install](https://docs.deepwisdom.ai/main/en/guide/get_started/installation.html#install-with-docker)
 
@@ -107,7 +95,9 @@ metagpt "Create a 2048 game"  # this will create a repo in ./workspace
 or use it as library
 
 ```python
-from metagpt.software_company import generate_repo, ProjectRepo
+from metagpt.software_company import generate_repo
+from metagpt.utils.project_repo import ProjectRepo
+
 repo: ProjectRepo = generate_repo("Create a 2048 game")  # or ProjectRepo("<path>")
 print(repo)  # it will print the repo structure with files
 ```
@@ -173,7 +163,7 @@ We will respond to all questions within 2-3 business days.
 
 To stay updated with the latest research and development, follow [@MetaGPT_](https://twitter.com/MetaGPT_) on Twitter. 
 
-To cite [MetaGPT](https://openreview.net/forum?id=VtmBAGCN7o) or [Data Interpreter](https://arxiv.org/abs/2402.18679) in publications, please use the following BibTeX entries.
+To cite [MetaGPT](https://openreview.net/forum?id=VtmBAGCN7o) in publications, please use the following BibTeX entries.   
 
 ```bibtex
 @inproceedings{hong2024metagpt,
@@ -183,54 +173,6 @@ To cite [MetaGPT](https://openreview.net/forum?id=VtmBAGCN7o) or [Data Interpret
       year={2024},
       url={https://openreview.net/forum?id=VtmBAGCN7o}
 }
-@misc{teng2025atom,
-      title={Atom of Thoughts for Markov LLM Test-Time Scaling}, 
-      author={Fengwei Teng and Zhaoyang Yu and Quan Shi and Jiayi Zhang and Chenglin Wu and Yuyu Luo},
-      year={2025},
-      eprint={2502.12018},
-      archivePrefix={arXiv},
-      primaryClass={cs.CL},
-      url={https://arxiv.org/abs/2502.12018}, 
-}
-@misc{xiang2025self,
-      title={Self-Supervised Prompt Optimization}, 
-      author={Jinyu Xiang and Jiayi Zhang and Zhaoyang Yu and Fengwei Teng and Jinhao Tu and Xinbing Liang and Sirui Hong and Chenglin Wu and Yuyu Luo},
-      year={2025},
-      eprint={2502.06855},
-      archivePrefix={arXiv},
-      primaryClass={cs.CL},
-      url={https://arxiv.org/abs/2502.06855}, 
-}
-@inproceedings{wang2025fact,
-      title={FACT: Examining the Effectiveness of Iterative Context Rewriting for Multi-fact Retrieval}, 
-      author={Jinlin Wang and Suyuchen Wang and Ziwen Xia and Sirui Hong and Yun Zhu and Bang Liu and Chenglin Wu},
-      booktitle={The 2025 Annual Conference of the Nations of the Americas Chapter of the ACL},
-      year={2025},
-      url={https://openreview.net/forum?id=VXOircx5h3}
-}
-@misc{chi2024sela,
-      title={SELA: Tree-Search Enhanced LLM Agents for Automated Machine Learning}, 
-      author={Yizhou Chi and Yizhang Lin and Sirui Hong and Duyi Pan and Yaying Fei and Guanghao Mei and Bangbang Liu and Tianqi Pang and Jacky Kwok and Ceyao Zhang and Bang Liu and Chenglin Wu},
-      year={2024},
-      eprint={2410.17238},
-      archivePrefix={arXiv},
-      primaryClass={cs.AI},
-      url={https://arxiv.org/abs/2410.17238}, 
-}
-@inproceedings{zhang2025aflow,
-      title={{AF}low: Automating Agentic Workflow Generation},
-      author={Jiayi Zhang and Jinyu Xiang and Zhaoyang Yu and Fengwei Teng and Xiong-Hui Chen and Jiaqi Chen and Mingchen Zhuge and Xin Cheng and Sirui Hong and Jinlin Wang and Bingnan Zheng and Bang Liu and Yuyu Luo and Chenglin Wu},
-      booktitle={The Thirteenth International Conference on Learning Representations},
-      year={2025},
-      url={https://openreview.net/forum?id=z5uVAKwmjf}
-}
-@misc{hong2024data,
-      title={Data Interpreter: An LLM Agent For Data Science}, 
-      author={Sirui Hong and Yizhang Lin and Bang Liu and Bangbang Liu and Binhao Wu and Danyang Li and Jiaqi Chen and Jiayi Zhang and Jinlin Wang and Li Zhang and Lingyao Zhang and Min Yang and Mingchen Zhuge and Taicheng Guo and Tuo Zhou and Wei Tao and Wenyi Wang and Xiangru Tang and Xiangtao Lu and Xiawu Zheng and Xinbing Liang and Yaying Fei and Yuheng Cheng and Zongze Xu and Chenglin Wu},
-      year={2024},
-      eprint={2402.18679},
-      archivePrefix={arXiv},
-      primaryClass={cs.AI},
-      url={https://arxiv.org/abs/2402.18679}, 
-}
 ```
+
+For more work, please refer to [Academic Work](academic_work.md).

academic_work.md

@@ -0,0 +1,60 @@
+```bibtex
+@inproceedings{hong2024metagpt,
+      title={Meta{GPT}: Meta Programming for A Multi-Agent Collaborative Framework},
+      author={Sirui Hong and Mingchen Zhuge and Jonathan Chen and Xiawu Zheng and Yuheng Cheng and Jinlin Wang and Ceyao Zhang and Zili Wang and Steven Ka Shing Yau and Zijuan Lin and Liyang Zhou and Chenyu Ran and Lingfeng Xiao and Chenglin Wu and J{\"u}rgen Schmidhuber},
+      booktitle={The Twelfth International Conference on Learning Representations},
+      year={2024},
+      url={https://openreview.net/forum?id=VtmBAGCN7o}
+}
+
+@misc{teng2025atom,
+      title={Atom of Thoughts for Markov LLM Test-Time Scaling}, 
+      author={Fengwei Teng and Zhaoyang Yu and Quan Shi and Jiayi Zhang and Chenglin Wu and Yuyu Luo},
+      year={2025},
+      eprint={2502.12018},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL},
+      url={https://arxiv.org/abs/2502.12018}, 
+}
+@misc{xiang2025self,
+      title={Self-Supervised Prompt Optimization}, 
+      author={Jinyu Xiang and Jiayi Zhang and Zhaoyang Yu and Fengwei Teng and Jinhao Tu and Xinbing Liang and Sirui Hong and Chenglin Wu and Yuyu Luo},
+      year={2025},
+      eprint={2502.06855},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL},
+      url={https://arxiv.org/abs/2502.06855}, 
+}
+@inproceedings{wang2025fact,
+      title={FACT: Examining the Effectiveness of Iterative Context Rewriting for Multi-fact Retrieval}, 
+      author={Jinlin Wang and Suyuchen Wang and Ziwen Xia and Sirui Hong and Yun Zhu and Bang Liu and Chenglin Wu},
+      booktitle={The 2025 Annual Conference of the Nations of the Americas Chapter of the ACL},
+      year={2025},
+      url={https://openreview.net/forum?id=VXOircx5h3}
+}
+@misc{chi2024sela,
+      title={SELA: Tree-Search Enhanced LLM Agents for Automated Machine Learning}, 
+      author={Yizhou Chi and Yizhang Lin and Sirui Hong and Duyi Pan and Yaying Fei and Guanghao Mei and Bangbang Liu and Tianqi Pang and Jacky Kwok and Ceyao Zhang and Bang Liu and Chenglin Wu},
+      year={2024},
+      eprint={2410.17238},
+      archivePrefix={arXiv},
+      primaryClass={cs.AI},
+      url={https://arxiv.org/abs/2410.17238}, 
+}
+@inproceedings{zhang2025aflow,
+      title={{AF}low: Automating Agentic Workflow Generation},
+      author={Jiayi Zhang and Jinyu Xiang and Zhaoyang Yu and Fengwei Teng and Xiong-Hui Chen and Jiaqi Chen and Mingchen Zhuge and Xin Cheng and Sirui Hong and Jinlin Wang and Bingnan Zheng and Bang Liu and Yuyu Luo and Chenglin Wu},
+      booktitle={The Thirteenth International Conference on Learning Representations},
+      year={2025},
+      url={https://openreview.net/forum?id=z5uVAKwmjf}
+}
+@misc{hong2024data,
+      title={Data Interpreter: An LLM Agent For Data Science}, 
+      author={Sirui Hong and Yizhang Lin and Bang Liu and Bangbang Liu and Binhao Wu and Danyang Li and Jiaqi Chen and Jiayi Zhang and Jinlin Wang and Li Zhang and Lingyao Zhang and Min Yang and Mingchen Zhuge and Taicheng Guo and Tuo Zhou and Wei Tao and Wenyi Wang and Xiangru Tang and Xiangtao Lu and Xiawu Zheng and Xinbing Liang and Yaying Fei and Yuheng Cheng and Zongze Xu and Chenglin Wu},
+      year={2024},
+      eprint={2402.18679},
+      archivePrefix={arXiv},
+      primaryClass={cs.AI},
+      url={https://arxiv.org/abs/2402.18679}, 
+}
+```

config/config2.example.yaml

@@ -20,6 +20,37 @@ embedding:
   embed_batch_size: 100
   dimensions: # output dimension of embedding model
 
+# Role's custom configuration
+roles:
+  - role: "ProductManager" # role's className or role's role_id
+    llm:
+      api_type: "openai"  # or azure / ollama / open_llm etc. Check LLMType for more options
+      base_url: "YOUR_BASE_URL"
+      api_key: "YOUR_API_KEY"
+      proxy: "YOUR_PROXY"  # for LLM API requests
+      model: "gpt-4-turbo-1106"
+  - role: "Architect"
+    llm:
+      api_type: "openai"  # or azure / ollama / open_llm etc. Check LLMType for more options
+      base_url: "YOUR_BASE_URL"
+      api_key: "YOUR_API_KEY"
+      proxy: "YOUR_PROXY"  # for LLM API requests
+      model: "gpt-35-turbo"
+  - role: "ProjectManager"
+    llm:
+      api_type: "azure"
+      base_url: "YOUR_BASE_URL"
+      api_key: "YOUR_API_KEY"
+      api_version: "YOUR_API_VERSION"
+      model: "gpt-4-1106"
+  - role: "Engineer"
+    llm:
+      api_type: "azure"
+      base_url: "YOUR_BASE_URL"
+      api_key: "YOUR_API_KEY"
+      api_version: "YOUR_API_VERSION"
+      model: "gpt-35-turbo-1106"
+
 repair_llm_output: true  # when the output is not a valid json, try to repair it
 
 proxy: "YOUR_PROXY"  # for tools like requests, playwright, selenium, etc.
@@ -50,6 +81,21 @@ s3:
   secure: false
   bucket: "test"
 
+exp_pool:
+  enabled: false
+  enable_read: false 
+  enable_write: false
+  persist_path: .chroma_exp_data # The directory.
+  retrieval_type: bm25 # Default is `bm25`, can be set to `chroma` for vector storage, which requires setting up embedding.
+  use_llm_ranker: true # Default is `true`, it will use LLM Reranker to get better result.
+  collection_name: experience_pool  # When `retrieval_type` is `chroma`, `collection_name` is the collection name in chromadb.
+
+role_zero:
+  enable_longterm_memory: false # Whether to use long-term memory. Default is `false`.
+  longterm_memory_persist_path: .role_memory_data # The directory to save data.
+  memory_k: 200 # The capacity of short-term memory.
+  similarity_top_k: 5 # The number of long-term memories to retrieve.
+  use_llm_ranker: false # Whether to use LLM Reranker to get better result. Default is `false`.
 
 azure_tts_subscription_key: "YOUR_SUBSCRIPTION_KEY"
 azure_tts_region: "eastus"
@@ -81,5 +127,3 @@ models:
 #    # timeout: 600 # Optional. If set to 0, default value is 300.
 #    # Details: https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/
 #    pricing_plan: "" # Optional. Use for Azure LLM when its model name is not the same as OpenAI's
-
-agentops_api_key: "YOUR_AGENTOPS_API_KEY" # get key from https://app.agentops.ai/settings/projects

config/vault.example.yaml

@@ -0,0 +1,48 @@
+# Usage:
+# 1. Get value.
+# >>> from metagpt.tools.libs.env import get_env
+# >>> access_token = await get_env(key="access_token", app_name="github")
+# >>> print(access_token)
+# YOUR_ACCESS_TOKEN
+#
+# 2. Get description for LLM understanding.
+# >>> from metagpt.tools.libs.env import get_env_description
+# >>> descriptions = await get_env_description
+# >>> for k, desc in descriptions.items():
+# >>>     print(f"{key}:{desc}")
+# await get_env(key="access_token", app_name="github"):Get github access token
+# await get_env(key="access_token", app_name="gitlab"):Get gitlab access token
+# ...
+
+vault:
+  github:
+    values:
+      access_token: "YOUR_ACCESS_TOKEN"
+    descriptions:
+      access_token: "Get github access token"
+  gitlab:
+    values:
+      access_token: "YOUR_ACCESS_TOKEN"
+    descriptions:
+      access_token: "Get gitlab access token"
+  iflytek_tts:
+    values:
+      api_id: "YOUR_APP_ID"
+      api_key: "YOUR_API_KEY"
+      api_secret: "YOUR_API_SECRET"
+    descriptions:
+      api_id: "Get the API ID of IFlyTek Text to Speech"
+      api_key: "Get the API KEY of IFlyTek Text to Speech"
+      api_secret: "Get the API SECRET of IFlyTek Text to Speech"
+  azure_tts:
+    values:
+      subscription_key: "YOUR_SUBSCRIPTION_KEY"
+      region: "YOUR_REGION"
+    descriptions:
+      subscription_key: "Get the subscription key of Azure Text to Speech."
+      region: "Get the region of Azure Text to Speech."
+  default: # All key-value pairs whose app name is an empty string are placed below
+    values:
+      proxy: "YOUR_PROXY"
+    descriptions:
+      proxy: "Get proxy for tools like requests, playwright, selenium, etc."

examples/cr.py

@@ -0,0 +1,15 @@
+import fire
+
+from metagpt.roles.di.engineer2 import Engineer2
+from metagpt.tools.libs.cr import CodeReview
+
+
+async def main(msg):
+    role = Engineer2(tools=["Plan", "Editor:write,read", "RoleZero", "ValidateAndRewriteCode", "CodeReview"])
+    cr = CodeReview()
+    role.tool_execution_map.update({"CodeReview.review": cr.review, "CodeReview.fix": cr.fix})
+    await role.run(msg)
+
+
+if __name__ == "__main__":
+    fire.Fire(main)