Repository navigation
Expand file tree
/
Copy pathplugin.json
More file actions
111 lines (111 loc) · 8.18 KB
/
Copy pathplugin.json
File metadata and controls
111 lines (111 loc) · 8.18 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
{
"$schema": "https://agent-plugins.org/schemas/1.0.0/plugin.schema.json",
"name": "fabric",
"version": "1.0.0",
"description": "Fabric is a personal AI workspace where everything you save, write, capture, or record lives in one intelligent space. Notes, files, PDFs, bookmarks, voice recordings, meeting transcripts, web clippings, photos, and more – all searchable by meaning, not just keywords.\n\nThe Fabric MCP connector gives ChatGPT direct access to your entire Fabric library. Ask ChatGPT to search your notes, find a PDF you saved last month, pull up highlights from a book, or reference a meeting transcript – without leaving the conversation.\n\nWhat you can do:\n\nSearch across everything in your Fabric library by meaning. Ask \"what were my notes on pricing strategy\" and get results from across notes, documents, bookmarks, and recordings.\n\nReference your own materials in conversations with ChatGPT. Instead of pasting content manually, ChatGPT can pull directly from your saved research, course materials, client files, or personal notes.\n\nBuild on your own knowledge. ChatGPT's intelligence combined with your personal library means answers grounded in your actual work – not generic internet results.\n\nFabric connects to all your favourite apps and tools – Google Drive, email, Linear, Google Calendar, Slack, and more – which feed into the same searchable library that ChatGPT can access.\n\nWhether you're a student studying from lecture recordings and textbook PDFs, a researcher working across hundreds of papers, a freelancer managing client projects, or a founder keeping track of competitive research – Fabric gives ChatGPT the context it needs to help you with your actual work.",
"author": {
"name": "Fabric",
"email": "team@fabric.so",
"url": "https://fabric.so"
},
"homepage": "https://fabric.so",
"keywords": [
"fabric",
"productivity",
"collaboration",
"notes",
"docs",
"wiki",
"projects",
"pdfs",
"photos",
"knowledge management",
"file storage",
"bookmarks",
"voice recordings",
"meeting transcripts"
],
"extensions": {
"com.openai": {
"interface": {
"displayName": "Fabric",
"shortDescription": "Ask your knowledge base",
"longDescription": "Fabric is a personal AI workspace where everything you save, write, capture, or record lives in one intelligent space. Notes, files, PDFs, bookmarks, voice recordings, meeting transcripts, web clippings, photos, and more – all searchable by meaning, not just keywords.\n\nThe Fabric MCP connector gives ChatGPT direct access to your entire Fabric library. Ask ChatGPT to search your notes, find a PDF you saved last month, pull up highlights from a book, or reference a meeting transcript – without leaving the conversation.\n\nWhat you can do:\n\nSearch across everything in your Fabric library by meaning. Ask \"what were my notes on pricing strategy\" and get results from across notes, documents, bookmarks, and recordings.\n\nReference your own materials in conversations with ChatGPT. Instead of pasting content manually, ChatGPT can pull directly from your saved research, course materials, client files, or personal notes.\n\nBuild on your own knowledge. ChatGPT's intelligence combined with your personal library means answers grounded in your actual work – not generic internet results.\n\nFabric connects to all your favourite apps and tools – Google Drive, email, Linear, Google Calendar, Slack, and more – which feed into the same searchable library that ChatGPT can access.\n\nWhether you're a student studying from lecture recordings and textbook PDFs, a researcher working across hundreds of papers, a freelancer managing client projects, or a founder keeping track of competitive research – Fabric gives ChatGPT the context it needs to help you with your actual work.",
"developerName": "Fabric",
"category": "Productivity",
"capabilities": ["Read", "Write"],
"websiteURL": "https://fabric.so",
"privacyPolicyURL": "https://fabric.so/info/privacy",
"termsOfServiceURL": "https://fabric.so/info/tos",
"supportURL": "https://fabric.so/info/contact",
"defaultPrompt": [
"Save that idea to my Fabric.",
"Search my Fabric for everything I've saved about this topic.",
"Summarise last week's meeting notes from Fabric."
],
"brandColor": "#000000",
"brancColorDark": "#FFFFFF",
"composerIcon": "./assets/composer-icon.png",
"logo": "./assets/logo.svg"
},
"mcpServers": "./mcp.json",
"publication": {
"countries": [],
"release_notes": "First release of the Fabric MCP plugin for ChatGPT.\n\nThe Fabric MCP connector gives ChatGPT direct access to your entire Fabric library. Ask ChatGPT to search your notes, find a PDF you saved last month, pull up highlights from a book, or reference a meeting transcript – without leaving the conversation."
},
"review": {
"test_cases": {
"positive": [
{
"description": "User searches their Fabric workspace for previously saved content by keyword",
"expected_behavior": "A list of matching saved links, files, or notes from the user's Fabric workspace that relate to machine learning, returned with titles and metadata",
"tools_triggered": "fabric-find-items",
"prompt": "Find my saved articles about machine learning in Fabric"
},
{
"description": "User asks to create a new note with a title and body content",
"expected_behavior": "A new note is created in Fabric with the specified title and markdown body. Confirmation returned with the note's title or ID",
"tools_triggered": "fabric-create-note",
"prompt": "Create a note in Fabric called 'Meeting Notes – Sept 11' with this summary: discussed roadmap priorities, assigned Q4 OKRs, flagged the API migration deadline"
},
{
"description": "User adds a new task to their Fabric task list",
"expected_behavior": "A new task is created in Fabric with the given description. Confirmation returned with the task details",
"tools_triggered": "fabric-create-task",
"prompt": "Add a task to Fabric: Review the Q3 analytics dashboard by Friday"
},
{
"description": "User stores a personal preference or fact as a Fabric memory",
"expected_behavior": "A new memory object is created in Fabric capturing the user's stated preferences. Confirmation returned",
"tools_triggered": "fabric-create-memory",
"prompt": "Remember on Fabric that my preferred stack is Python + FastAPI and I work at Acme Corp"
},
{
"description": "User saves a URL as a bookmark and moves it into an existing folder",
"expected_behavior": "A bookmark is created for the URL, then moved into the user's Research folder. Confirmation of both actions returned",
"tools_triggered": "fabric-find-items, fabric-create-bookmark, fabric-move-items",
"prompt": "Save this link to Fabric: https://example.com/design-systems-guide and put it in my Research folder"
}
],
"negative": [
{
"description": "User asks about fabric as a physical material — no Fabric app functionality should trigger",
"prompt": "What's the best fabric for a summer dress?"
},
{
"description": "User asks a general technical knowledge question about the MCP protocol itself, not about using Fabric's MCP tools",
"prompt": "How do I build my own MCP server from scratch?"
},
{
"description": "User asks to save content to a different service, not Fabric",
"prompt": "Save this document to my Google Drive Content"
}
]
},
"demo_recording_url": "https://youtu.be/tWXfGcQP9EQ",
"commerce": false,
"commerce_description": "This plugin does not sell products or process payments."
}
}
}
}