{"id":15017168,"url":"https://github.com/aurelienstebe/gladdis","last_synced_at":"2025-08-02T06:10:42.722Z","repository":{"id":158917266,"uuid":"608325644","full_name":"AurelienStebe/Gladdis","owner":"AurelienStebe","description":"Gladdis (Generative Language Artificial Dedicated \u0026 Diligent Intelligence System) - it's an AI chatbot.","archived":false,"fork":false,"pushed_at":"2025-03-26T14:10:03.000Z","size":1532,"stargazers_count":44,"open_issues_count":2,"forks_count":2,"subscribers_count":2,"default_branch":"main","last_synced_at":"2025-03-31T17:07:54.177Z","etag":null,"topics":["ai","artificial-intelligence","chatbot","gladdis","obsidian","openai"],"latest_commit_sha":null,"homepage":"","language":"TypeScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/AurelienStebe.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":".github/CODE_OF_CONDUCT.md","threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2023-03-01T19:36:58.000Z","updated_at":"2025-02-14T14:05:03.000Z","dependencies_parsed_at":"2024-11-14T17:02:33.633Z","dependency_job_id":"b3877b15-65ee-468f-92d2-076a91546b30","html_url":"https://github.com/AurelienStebe/Gladdis","commit_stats":{"total_commits":73,"total_committers":2,"mean_commits":36.5,"dds":0.0821917808219178,"last_synced_commit":"a96bb08ef490d6c5970189a5213e2e1299cef379"},"previous_names":[],"tags_count":10,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/AurelienStebe%2FGladdis","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/AurelienStebe%2FGladdis/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/AurelienStebe%2FGladdis/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/AurelienStebe%2FGladdis/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/AurelienStebe","download_url":"https://codeload.github.com/AurelienStebe/Gladdis/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":252998365,"owners_count":21838003,"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","artificial-intelligence","chatbot","gladdis","obsidian","openai"],"created_at":"2024-09-24T19:49:59.312Z","updated_at":"2025-05-08T04:25:47.270Z","avatar_url":"https://github.com/AurelienStebe.png","language":"TypeScript","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cp align=\"center\"\u003e\n    \u003cimg height=\"230\" width=\"320\" alt=\"Gladdis Logo\" src=\"https://raw.githubusercontent.com/AurelienStebe/Gladdis/main/gladdis.svg\"\u003e\n\u003c/p\u003e\n\n## Description\n\n[Obsidian](https://obsidian.md) exists to help you build a second brain. Gladdis exists to give that brain a pen (let it do the writing for once), ears (talking is easier than typing) and multiple split personalities (give your system prompts names). Further plans for v1.0 include: a brush (let it paint like an artist), a voice (listening is easier than reading), an internal monologue (together we are more than our sum) and the ability to execute code (what could possibly go wrong ?).\n\n### Current Features\n\n- **Native Markdown Syntax**: the conversation window is any open Markdown file, the streaming LLM response is appended.\n- **Whisper Transcription**: audio files WikiLinks are auto-transcribed, use the Obsidian \"audio recorder\" core plugin.\n- **Document Loading**: text files WikiLinks are auto-loaded, just drag and drop files into the conversation window.\n- **PDF \u0026 Web Support**: load PDF WikiLinks just like text files, load web pages external links as well using `\u003cURL\u003e`.\n- **Image Vision AIs**: send images to vision-supporting LLMs using local WikiLinks or Markdown external web links.\n- **System Prompts**: name your system prompts, create personalities and use FrontMatter to choose whom to talk to.\n\n## Plugin Demo Vault\n\nDownload the [Plugin Demo Vault in ZIP or TGZ](https://github.com/AurelienStebe/Gladdis/releases/latest) to try it out !\n\n\u003cp align=\"center\"\u003e\n    \u003cimg style=\"margin-top: 1em;\" alt=\"Gladdis Demo Vault\" src=\"https://raw.githubusercontent.com/AurelienStebe/Gladdis/main/picdemo.png\"\u003e\n\u003c/p\u003e\n\n## Documentation\n\n### Obsidian Commands\n\nGladdis provides 3 commands, map them to hotkeys in the settings or create buttons with the [Commander plugin](https://github.com/phibr0/obsidian-commander).\n\n- **Chat with Gladdis**: the main command to perform the full processing, including the call to the LLM model.\n- **Process the Content**: to check that all links and transcriptions are correct, and get a full token count.\n- **Process the Prompt (or Selection)**: to check just the prompt or selection and get a specific token count.\n\n### Settings / Options\n\nThe settings panel lets you specify the plugin data folder, your OpenAI API key, and the FrontMatter defaults.\n\n- **Data Root Path**: Gladdis will search for config files in the `configs` subfolder, and will write the chat and call logs in the `history` subfolder.\n- **OpenAI Secret Key**: create an OpenAI API account, [generate an API key](https://platform.openai.com/account/api-keys) and paste it in the settings panel.\n- **FrontMatter Defaults**: adjust the various defaults for the Gladdis and Whisper options.\n\n\u003cp align=\"center\"\u003e\n    \u003cimg style=\"margin-top: 2em;\" alt=\"The Settings Panel\" src=\"https://raw.githubusercontent.com/AurelienStebe/Gladdis/main/options.png\"\u003e\n\u003c/p\u003e\n\n### FrontMatter Config\n\nThe defaults from the settings panel can be overridden using FrontMatter in any conversation or config file.\n\n```yaml\n---\ngladdis:\n    config: Gladdis   # select which AI you talk to\n    model: gpt-4o     # choose the current AI model\n    server: http://   # use local or custom servers\n    temperature: 42   # test different temperatures\n    top_p_param: 88   # or various probability mass\nwhisper:\n    config: Whisper   # switch transcription config\n    model: whisper-1  # change the transcription AI\n    temperature: 24   # test different temperatures\n    echoOutput: true  # write transcription callout\n    deleteFile: false # cleanup after transcription\n---\n```\n\n### Markdown Syntax\n\nThe same familiar Obsidian syntax from your notes is used for the conversation history (it's all just text).\n\n```md\nJust use standard Markdown text for `User` messages and prompts.\nThe content of the Gladdis config file will be prepended to this.\nUse triple underscores to separate the sections of conversation.\n\n___\n\n__UserName:__ labels also start a new message with custom names.\n\n__Gladdis:__ Gladdis' label and `Assistant` are for AI messages.\n\n__System:__ is for System Prompts (the default in config files).\n\n___\n\n\u003e Quoted lines are ignored, error and token counter callouts too.\n\n[[path/file.md|WikiLinks to file]] and ![[audio_recording.mp3]] will be parsed.\nStatic Web pages can be loaded using angle brackets: \u003chttps://www.example.com\u003e.\n\nThe very last message is the prompt, the LLM response streams below.\n```\n\n### Available Models\n\nThe following [OpenAI models](https://platform.openai.com/docs/models) are supported via the paying API:\n\n| Label               | Tokens | Vision | Notes                                                                                          |\n| ------------------- | :----: | :----: | ---------------------------------------------------------------------------------------------- |\n| `o3-mini`           | 200 k  |        | OpenAI's latest reasoning model designed to excel at science, math, and coding tasks.          |\n| `o1-mini`           | 128 k  |        | OpenAI's first reasoning model designed to solve hard problems across domains.                 |\n| `gpt-4o`            | 128 k  |   ✅   | OpenAI's high-intelligence flagship model for complex, multi-step tasks.                       |\n| `gpt-4o-mini`       | 128 k  |   ✅   | OpenAI's affordable and intelligent small model for fast, lightweight tasks.                   |\n| `chatgpt-4o-latest` | 128 k  |   ✅   | Dynamic model continuously updated to the current version of GPT-4o in ChatGPT.                |\n| `gpt-4-turbo`       | 128 k  |   ✅   | The latest GPT-4 Turbo model with vision capabilities (GPT-4o is cheaper and faster).          |\n| `gpt-3.5-turbo`     |  16 k  |        | The latest GPT-3.5 Turbo model with higher accuracy (GPT-4o mini is cheaper and more capable). |\n\nOpen-Source models can be used locally via [Ollama](https://ollama.com/) or [LocalAI](https://localai.io/):\n\n| Label             | Tokens | Vision | Notes                                                                        |\n| ----------------- | :----: | :----: | ---------------------------------------------------------------------------- |\n| `llama3.3`        | 128 k  |        | Meta's new 70B model with similar performance to the Llama 3.1 405B model.   |\n| `llama3.2-vision` | 128 k  |   ✅   | Meta's image reasoning generative models in both 11B and 90B model sizes.    |\n| `llama3.2`        | 128 k  |        | Meta's new 1B and 3B models optimized for multilingual dialogue use cases.   |\n| `llama3.1`        | 128 k  |        | Meta's 8B, 70B and 405B models with overall stronger reasoning capabilities. |\n| `phi4`            |  16 k  |        | Microsoft's new Phi 4 \"Medium\" (14B) open model with a 16k context window.   |\n| `phi3.5-vision`   | 128 k  |   ✅   | Microsoft's Phi 3 and 3.5 \"Mini\" (3.8B) models with vision capabilities.     |\n| `phi3.5` / `phi3` | 128 k  |        | Microsoft's new Phi 3.5 \"Mini\" (3.8B) and Phi 3 \"Medium\" (14B) open models.  |\n| `qwen2.5`         |  32 k  |        | Alibaba Cloud's Qwen multilingual Chinese models ranging from 0.5B to 72B.   |\n| `gemma2`          |   8 k  |        | Google's Gemma 2B, 9B and 27B models featuring a brand new architecture.     |\n| `llava`           |   4 k  |   ✅   | Family of Large Language-and-Vision Assistant models between 7B and 34B.     |\n\n## Future Development\n\n- **Image \u0026 Audio Out**: create images with DALL·E, speak with ElevenLabs or other API accessible models.\n- **Code Execution**: add [Templater](https://github.com/SilentVoid13/Templater) code to your prompts, run any code the AI generates (are you sure ?!).\n- **AI-2-AI Prompt**: let your various AI configs talk to each other, create your own \"Council of Gladdises\".\n\n## Installation\n\n### From within Obsidian\n\n1. Open the \"Community plugins\" tab in the settings.\n2. Click the \"Browse\" button and search for \"Gladdis\".\n3. Click the \"Install\" button and enable the plugin.\n\n### With the BRAT plugin\n\n1. Install the [Beta Reviewers Auto-update Tool plugin](https://github.com/TfTHacker/obsidian42-brat).\n2. Follow the instructions for [\"Adding a beta plugin\"](https://tfthacker.com/brat-quick-guide#Adding+a+beta+plugin).\n\n### Manually from GitHub\n\n1. Download the `main.js` \u0026 `manifest.json` files from the [latest release](https://github.com/AurelienStebe/Gladdis/releases) into your vault's plugins folder: `\u003cvault\u003e/.obsidian/plugins/gladdis/`.\n\n## How to Contribute ?\n\nFeel free to open new issues or submit PRs, while the Gladdis community is growing.\n\n## About the Author\n\nGladdis is created by [Aurélien Stébé](https://github.com/AurelienStebe), a senior French freelance [Toptal Software Engineer](https://www.toptal.com/resume/aurelien-stebe).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Faurelienstebe%2Fgladdis","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Faurelienstebe%2Fgladdis","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Faurelienstebe%2Fgladdis/lists"}