{"id":13585014,"url":"https://github.com/oneil512/INSIGHT","last_synced_at":"2025-04-07T06:32:14.700Z","repository":{"id":152776133,"uuid":"625265280","full_name":"oneil512/INSIGHT","owner":"oneil512","description":"INSIGHT is an autonomous AI that can do medical research!","archived":false,"fork":false,"pushed_at":"2023-10-21T22:34:08.000Z","size":119,"stargazers_count":403,"open_issues_count":1,"forks_count":56,"subscribers_count":13,"default_branch":"master","last_synced_at":"2024-11-06T02:38:19.646Z","etag":null,"topics":["agent","ai","chatgpt","gpt","llm","medical","ml","python"],"latest_commit_sha":null,"homepage":"","language":"Python","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/oneil512.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":null,"security":null,"support":null,"governance":null}},"created_at":"2023-04-08T15:20:14.000Z","updated_at":"2024-10-21T16:13:39.000Z","dependencies_parsed_at":"2023-07-19T00:02:42.020Z","dependency_job_id":"5b2cda5f-23d2-4fbc-856f-ccb5e461c53b","html_url":"https://github.com/oneil512/INSIGHT","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/oneil512%2FINSIGHT","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/oneil512%2FINSIGHT/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/oneil512%2FINSIGHT/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/oneil512%2FINSIGHT/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/oneil512","download_url":"https://codeload.github.com/oneil512/INSIGHT/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247607274,"owners_count":20965942,"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":["agent","ai","chatgpt","gpt","llm","medical","ml","python"],"created_at":"2024-08-01T15:04:41.056Z","updated_at":"2025-04-07T06:32:14.690Z","avatar_url":"https://github.com/oneil512.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"# INSIGHT\n\nPlease visit https://insightai.dev/project for our managed solution with many more features!\n\nInsight is an autonomous AI that can do medical research. It has a boss agent that takes an objective and an executive summary of the tasks completed already and their results and creates a task list. A worker agent picks up a task from the list and completes it, saving the results to llama index. The boss gets informed of the results and changes/reprioritizes the task list. The workers can call into the pubmed and mygene APIs (more to come). The workers also get context from llama index to help complete their tasks.\n\nINSIGHT can also reload and continue runs, and also load any human readable data file and use it along side the other findings!\n\nYou can also load your llama Index database and talk to it, asking arbitrary questions about your data, by running `talk_to_index.py`\nYou will have to specify the path to your index in the bottom of the file. See the bottom of `talk_to_index.py` for an example.\n\nPlease reach out to me or contribute if this interests you :) My email is oneil512@umn.edu\n\n\n```mermaid\ngraph TB;\n\n    subgraph APIs;\n        API1[PUBMED API];\n        API2[MYGENE API];\n    end;\n\n    Boss((BOSS AGENT)) \u003c--\u003e GPT[LLM];\n    Llama[(LLAMA INDEX)] --\u003e|Summary of results| Boss;\n    Boss --\u003e|Create| Queue[TASK LIST];\n\n    Worker((WORKER AGENT)) \u003c--\u003e GPT;\n    Queue --\u003e |Pull| Worker;\n    Llama --\u003e|Context for task| Worker;\n    Worker --\u003e Result[Task Result];\n\n    Result --\u003e |Text| Llama;\n    Result --\u003e|Code| Executor{PYTHON EXECUTOR};\n\n    Executor --\u003e API1[PUBMED];\n    Executor --\u003e API2[MYGENE];\n    Executor --\u003e Execution[Execution Result];\n\n    Execution --\u003e Llama;\n\n    Llama \u003c--\u003e TalkToIndex[Talk To Index];\n\n    User{{User}} --\u003e|Query| TalkToIndex;\n    TalkToIndex --\u003e|Result| User;\n```\n\n## Getting Started\n\n1. Sign up for [OpenAI](https://platform.openai.com/signup)\n\n2. Expose the following environment variable\n    - OPENAI_API_KEY\n\n    OR\n\n    Add your api key to the config file. IF YOU DO THIS, DO NOT COMMIT THEM WITH ANY VERSION CONTROL SYSTEM!\n\n3. run `pip install -r requirements.txt`\n4. run `python main.py`\n\n\n## Output\n\nThe program saves the result from every task and adds it to the output directory `out`\n\nIt also creates a key findings markdown file over all results that distills the data via the following commands:\n\n* Give a brief high level summary of all the data.\n* Briefly list all the main points that the data covers.\n* Give all of the key insights about the data.\n* Generate several creative hypotheses given the data.\n* What are some high level research directions to explore further given the data?\n* Describe the key findings in great detail. Do not include filler words.\n\nArbitrary commands can be added. Open this in a markdown editor for the best experience.\n\nHere is an example output structure\n\n```\n.\n└── out  /\n    ├── Objective  /\n    │   ├── Task 1/\n    │   │   ├── Result 1/\n    │   │   │   ├── Raw Result\n    │   │   │   └── Vector Embedding of Result\n    │   │   ├── Result 2/\n    │   │   │   ├── Raw Result\n    │   │   │   └── Vector Embedding of Result\n    │   │   ├── .\n    │   │   ├── .\n    │   │   ├── Summary of task results\n    │   │   └── API Call (If task was an API call)\n    │   ├── Task 2\n    │   ├── .\n    │   ├── .\n    │   ├── .\n    │   └── Task N\n    └── key_findings.md\n```\n\n\nBE MINDFUL OF EXPENSES!!\n\nCurrently an execution for a few minutes should cost no more than a few cents. This will go up if you use a more powerful model like GPT-4","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Foneil512%2FINSIGHT","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Foneil512%2FINSIGHT","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Foneil512%2FINSIGHT/lists"}