{"id":13435489,"url":"https://github.com/litanlitudan/skyagi","last_synced_at":"2025-05-15T22:12:18.926Z","repository":{"id":154107966,"uuid":"626239608","full_name":"litanlitudan/skyagi","owner":"litanlitudan","description":"SkyAGI: Emerging human-behavior simulation capability in LLM","archived":false,"fork":false,"pushed_at":"2023-09-21T12:38:34.000Z","size":11261,"stargazers_count":780,"open_issues_count":40,"forks_count":56,"subscribers_count":18,"default_branch":"main","last_synced_at":"2025-05-13T14:18:17.816Z","etag":null,"topics":["ai-agent","aigc","langchain","language-model","llm"],"latest_commit_sha":null,"homepage":"https://skyagi.ai","language":"TypeScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/litanlitudan.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":"CODEOWNERS","security":null,"support":null,"governance":null,"roadmap":null,"authors":null}},"created_at":"2023-04-11T04:28:00.000Z","updated_at":"2025-05-09T04:03:19.000Z","dependencies_parsed_at":"2024-01-06T20:55:48.450Z","dependency_job_id":null,"html_url":"https://github.com/litanlitudan/skyagi","commit_stats":null,"previous_names":[],"tags_count":27,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/litanlitudan%2Fskyagi","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/litanlitudan%2Fskyagi/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/litanlitudan%2Fskyagi/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/litanlitudan%2Fskyagi/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/litanlitudan","download_url":"https://codeload.github.com/litanlitudan/skyagi/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":254430335,"owners_count":22069909,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai-agent","aigc","langchain","language-model","llm"],"created_at":"2024-07-31T03:00:36.121Z","updated_at":"2025-05-15T22:12:18.908Z","avatar_url":"https://github.com/litanlitudan.png","language":"TypeScript","funding_links":[],"categories":["LangChain Agents List","Game / Simulation","TypeScript","工具","Tools","Learning","前沿项目"],"sub_categories":["Open Source","代理","Agents","Repositories"],"readme":"# SkyAGI: Emerging human-behavior simulation capability in LLM\n\nhttps://github.com/litanlitudan/skyagi/assets/4970420/a584b364-e659-476d-a884-932b3c04df61\n\n\u003cp align=\"center\"\u003e\n    \u003ca href=\"https://pypi.org/project/skyagi/\"\u003e\n        \u003cimg alt=\"PyPI\" src=\"https://img.shields.io/pypi/v/skyagi?color=gree\"\u003e\n    \u003c/a\u003e\n\u003c/p\u003e\n\n`SkyAGI` is a python package that demonstrates LLM's emerging capability in simulating believable human behaviors.\nSpecifically, `SkyAGI` implements the idea of [Generative Agents](https://arxiv.org/abs/2304.03442) and delivers a role-playing game that creates a very interesting user experience.\n\nDifferent from previous AI based NPC systems, `SkyAGI`'s NPC generates very believable human responses.\nThe interesting observations in this demo show a huge potential for rethinking game development in many aspects, such as NPC script writing.\n\nTo demonstrate this, `SkyAGI` provides example characters from `The Big Bang Theory` and `The Avengers` as a starting point.\nUsers could also define customized characters by creating config json files like [customized_character.json](https://github.com/litanlitudan/skyagi/blob/main/examples/example_agent.json)\nFor details about the interesting observations, refer to the [observations section](https://github.com/litanlitudan/skyagi/#interesting-observations-in-this-demo).\n\n## Quick Start\n\nInstallation\n\n```sh\npip install --upgrade skyagi\n```\n\nOr\n\n```sh\nmake install\n```\n\nHow to run\n\n```sh\nexport OPENAI_API_KEY=\"...\"\nskyagi\n# or\nOPENAI_API_KEY=\"...\" skyagi\n```\n\nFor example if the OpenAI key is `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`, then the exact command would be the following\n\n```sh\n# make sure no quote around the token\nexport OPENAI_API_KEY=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\nskyagi\n# or\nOPENAI_API_KEY=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 skyagi\n```\n\nTo use example agent configs, download it from here: https://github.com/litanlitudan/skyagi/tree/main/examples\n(pip install doesn't contain the agent configuration)\n\nAn example agent configuration (Sheldon) looks something like this:\n\n```json\n{\n    \"name\": \"Sheldon\",\n    \"age\": 27,\n    \"personality\": \"Intelligent, rigid, socially challenged, quirky, and arrogant.\",\n    \"memories\": [\n        \"Sheldon is a theoretical physicist who works at Caltech.\",\n        \"Sheldon has an eidetic memory and is highly intelligent, but struggles with social skills and sarcasm.\",\n        ...\n        \"Knock, knock, knock, Penny - This is the specific knock that Sheldon uses when he visits Penny's apartment, which he repeats three times.\",\n        \"Bazinga! - This is Sheldon's catchphrase that he uses to indicate he was joking or playing a prank on someone.\"\n    ],\n    \"current_status\": \"Sheldon is at the Cheesecake Factory\"\n}\n```\n\n## Interesting observations in this demo\n\nHere is a screenshot of a live demo using The Big Bang Theory example.\n![demo](./images/demo.png)\nFrom the conversation, we can observe three interesting points that have not been widely seen in previous systems:\n\n1. Leonard remembered that Penny had asked him to persuade Sheldon to go for a hike, which shows the capability of some kind of memory.\n2. Leonard changed his mind after Sheldon whispered to him and even tried to convince Penny to join the scientific effort, which shows that the agents had meaningful progress in the story even without human intervention.\n3. All the responses are quite human-like. As a user, it's quite hard to tell whether it's actually an AI behind the responses.\n\n## References\n\n1. https://arxiv.org/abs/2304.03442\n2. https://python.langchain.com/en/latest/use_cases/agent_simulations/characters.html#create-a-generative-character\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flitanlitudan%2Fskyagi","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Flitanlitudan%2Fskyagi","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flitanlitudan%2Fskyagi/lists"}