# Deepractice/PromptX

PromptX · 领先的AI 智能体上下文平台 ｜ PromptX · Leading AI Agent Context Platform

Repository: https://github.com/Deepractice/PromptX
Canonical: https://ross.abutalabs.com/products/promptx
Homepage: https://promptx.deepractice.ai
Language: TypeScript
License: MIT
License Family: permissive
Last push: 2026-05-17T08:01:53+00:00

## Health v2 (maintenance only)
Score: 71/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 82, release rhythm 79, longevity 34
- inputs: {"age_days": 477, "days_push": 108, "days_rel": 143, "gap_med": 1, "n_releases_24m": 52}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3688, forks 309 (observed 2026-08-28T04:08:14.701934+00:00)

## What it is
PromptX is an AI Agent Context Platform built on the Model Context Protocol (MCP) that injects professional AI roles, a cognitive memory system, and sandboxed tools into MCP-enabled applications like Claude Desktop and Cursor. It ships as a desktop app, npm package, or Docker container running an MCP server on port 5203.

## Use cases
- give claude desktop persistent memory across conversations
- create custom expert personas for my ai assistant
- add professional ai roles to cursor
- run an mcp server with sandboxed tools for file and document operations
- build custom tools for ai agents without writing much code
- replace rag with semantic memory for my ai agent
- remember user context and learning progress in ai tutoring

## When to choose
- you use MCP-enabled apps like Claude Desktop or Cursor and want roles, memory, and tools
- you want engram-based cognitive memory instead of chunk-and-search RAG
- you need a sandboxed runtime for AI-executed tools defined in YAML

## When to avoid
- you need a hosted cloud solution with no local components
- your AI client does not support the Model Context Protocol
- you only need simple prompt templates without memory or tooling

## Facets
- artifact type: service
- maturity: active
- function: mcp, agent-framework, rag, chatbot, prompt-engineering
- domain: artificial-intelligence, large-language-models, developer-tools
- platform: cross-platform, self-hosted
- tags: mcp-server, ai-roles, cognitive-memory, toolx, sandboxed-tools, engram-memory, claude-desktop, cursor, ai-agents, nodejs, docker, desktop

## Member repositories
- Deepractice/PromptX (main) score 71

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:14.701934+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:31:08.427516+00:00, confidence not recorded.
  - readme: https://github.com/Deepractice/PromptX (fetched 2026-08-28T04:08:14.701934+00:00, sha f1dc99e4763e)
  - homepage: https://promptx.deepractice.ai (fetched 2026-08-29T09:25:18.417602+00:00, sha 0c8ba04eebe8)
  - site_page: https://promptx.deepractice.ai/docs (fetched 2026-08-29T09:25:18.420030+00:00, sha 9531253a0d47)
  - site_page: https://promptx.deepractice.ai/docs/quick-start (fetched 2026-08-29T09:25:18.421707+00:00, sha a43ab241e492)
  - site_page: https://deepractice.ai (fetched 2026-08-29T09:25:18.423354+00:00, sha 81821f248d88)
- Data as of 2026-08-30T08:39:29.467469+00:00.