Reading
mcp-crew is a Claude plugin based on the MCP protocol. It enables Claude to switch between 8 professional roles (including CEO, CFO, CTO, and Product Manager) via a persona mechanism, with no additional API call costs. It supports memory integration and multi-role workflows.
Section 01
mcp-crew is a Claude plugin based on the MCP protocol. It allows Claude to switch between 8 professional roles (such as CEO, CFO, CTO, and Product Manager) via a persona mechanism, with no additional API call costs. It supports memory integration and multi-role workflows, helping users fully leverage the value of their existing Claude subscription.
Section 02
In traditional multi-agent architectures, each agent requires an independent LLM API call, so costs grow linearly with the number of roles. For example, a task with 3 roles needs 3 calls; when the number increases to 8 or more, costs can quickly get out of control.
Section 03
mcp-crew enables Claude to switch roles internally (not as independent agents) via the MCP protocol, using system prompts to achieve professional role transitions. Currently, it offers 8 built-in roles:
Section 04
Each role is equipped with complete professional tools: domain frameworks (e.g., OKR for CEO, SaaS metrics for CFO), few-shot examples, anti-pattern warnings, chain-of-thought processes, and cross-role handover guidance. It also supports integration with StudioMeyer Memory, where each role automatically loads business context (e.g., CFO knows pricing data, CEO understands strategic goals) to provide personalized recommendations.
Section 05
It supports multi-role workflows, such as strategic review (CEO→CFO→CTO) and content production (CMO→Analyst→Creative). It provides 10 MCP tools: crew_guide (guide), crew_list (role list), crew_activate (activate role), crew_deactivate (return to default), crew_status (status), crew_feedback (rating), crew_create (custom role), crew_delete (delete role), crew_workflow_list (workflow list), crew_workflow_run (run workflow).
Section 06
You can create custom roles (e.g., legal advisor) and save them to the local directory for immediate effect. Comparison with competitors:
| Feature | mcp-crew | DollhouseMCP | Ruflo |
|---|---|---|---|
| Number of roles | 8 (with frameworks) | 150+ (general) | 60+ (group) |
| Memory integration | Yes | No | Proprietary |
| Additional API cost | 0 | 0 | $$$ |
| Framework support | Multiple (RICE/DORA, etc.) | None | Templates |
| Setup | 1 URL | npm installation | npm + wizard |
| The advantage of mcp-crew lies in deep professional frameworks rather than the number of roles. |
Section 07
Usage methods:
https://crew.studiomeyer.io/mcp and log in with your email to use.Keep going with more reads from the same topic.
Splinter is a minimalist, high-performance key-value (KV) and vector storage system enabling zero-latency inter-process communication via shared memory and atomic operations. With only 766 lines of core code, it supports millions of operations per second and 768-dimensional vector storage, offering a new architectural approach for local LLM inference and data-intensive applications.
LiteMind is a Python framework that provides developers with a unified API to integrate mainstream LLM providers such as OpenAI, Anthropic, Google Gemini, and Ollama. It supports multimodal input/output, tool calling, RAG enhancement, and agent construction.
OmniRoute is an open-source AI gateway that integrates over 67 large model providers through a single OpenAI-compatible endpoint. It offers intelligent routing, load balancing, automatic degradation, and quota management, helping developers get the most available AI services at the lowest cost.
This article introduces an open-source multimodal RAG framework based on Google Gemini Embedding 2, which can uniformly handle embedding and retrieval of four media types: text, images, videos, and audio. Combined with Supabase pgvector vector database and OpenRouter large language models, it provides a complete production-level retrieval-augmented generation pipeline.