convez.ai
MCP ServerIntegração & APIsAvançado
entire-vc

EVC Mesh MCP Server

entire-vc/evc-mesh-mcp

Conecte agentes de IA ao EVC Mesh para gestão de tarefas colaborativas

Ver no GitHub
InicialPluginScore · de 0 a 100

Servidor MCP que integra agentes de IA (Claude, Cursor, Cline) com a plataforma EVC Mesh para gerenciamento de tarefas, memória persistente e coordenação multi-agente. Oferece dois perfis de ferramentas (core com 20 tools e full com 45 tools) para otimizar uso de contexto, com suporte a transporte stdio e SSE.

Instalação

$npx convez-ai entire-vc/evc-mesh-mcp

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
1
Forks
0
Acompanhando
1
Saúde
50%
Último commit
há 2 dias
Versão
v0.1.8
Issues abertas
0
Contribuidores
5

Casos de uso

  • Coordenação de múltiplos agentes de IA em projetos compartilhados
  • Gerenciamento de tarefas com memória persistente entre sessões
  • Automação de fluxos de trabalho com colaboração humano-IA
  • Integração de assistentes de código (Claude, Cursor, Cline) em plataformas de gerenciamento de projetos

Veredito

MCP server robusto e bem documentado para teams que coordenam múltiplos agentes IA. Demanda setup técnico (Go, chave API, instância Mesh própria), mas resolve genuinamente o problema de persistência e coordenação entre agentes. Ideal para automação enterprise.

README

A tradução para o português deste README ainda está na fila.

EVC Mesh MCP Server

Install via Spark

Model Context Protocol (MCP) server for EVC Mesh — a task management platform for coordinating humans and AI agents.

Connects AI agents (Claude Code, Cursor, Cline, OpenClaw, etc.) to EVC Mesh via MCP tools for task management, persistent memory, event publishing, and multi-agent coordination.

Prerequisites

  • Go 1.22+
  • Running EVC Mesh instance
  • Agent registered in Mesh with an API key (agk_...)

Installation

go install github.com/entire-vc/evc-mesh-mcp@latest

Or build from source:

git clone https://github.com/entire-vc/evc-mesh-mcp.git
cd evc-mesh-mcp
go build -o evc-mesh-mcp .

Tool Profiles

The MCP server supports two profiles to optimize context window usage:

ProfileToolsContext overheadBest for
core20~6K tokens (3% of 200K)Claude Code, Cursor, small-context models
full45~14K tokens (7% of 200K)Power users, automation agents, admin ops

Set via MESH_MCP_PROFILE environment variable. Default: full.

Configuration

VariableRequiredDefaultDescription
MESH_API_URLYeshttp://localhost:8005Base URL of the Mesh API
MESH_AGENT_KEYYes (stdio)—Agent API key (agk_...)
MESH_MCP_PROFILENofullTool profile: core or full
MESH_MCP_TRANSPORTNostdioTransport mode: stdio or sse
MESH_MCP_HOSTNo0.0.0.0SSE server bind host
MESH_MCP_PORTNo8081SSE server bind port

Claude Code (stdio mode)

Add to your project's .mcp.json:

{
  "mcpServers": {
    "evc-mesh": {
      "command": "evc-mesh-mcp",
      "env": {
        "MESH_API_URL": "https://your-mesh-instance.example.com",
        "MESH_AGENT_KEY": "agk_your-workspace_your-key",
        "MESH_MCP_PROFILE": "core"
      }
    }
  }
}

Cursor

Add to Cursor MCP settings (Settings → MCP Servers):

{
  "evc-mesh": {
    "command": "evc-mesh-mcp",
    "env": {
      "MESH_API_URL": "https://your-mesh-instance.example.com",
      "MESH_AGENT_KEY": "agk_your-workspace_your-key",
      "MESH_MCP_PROFILE": "core"
    }
  }
}

SSE Mode (multi-agent, shared server)

For connecting multiple agents through a shared MCP endpoint:

MESH_API_URL=https://your-mesh-instance.example.com \
MESH_MCP_PORT=8081 \
evc-mesh-mcp --transport sse

SSE mode serves two profiles simultaneously on different paths:

PathProfileDescription
/sse + /messagefullAll 45 tools (backward compatible)
/core/sse + /core/messagecore20 essential tools

Authentication per connection via:

  • Authorization: Bearer agk_... header
  • X-Agent-Key: agk_... header
  • ?agent_key=agk_... query parameter

Agent Context Protocol (ACP)

At session start, follow these 5 steps in order:

1. heartbeat(status="online")              → register as alive
2. get_project_knowledge(project_id)       → load accumulated decisions & conventions
3. get_my_rules(project_id)                → understand constraints
4. get_context(project_id)                 → see recent activity + project knowledge
5. get_my_tasks()                          → check assigned work

At session end:

publish_event(type="summary", memory={persist: true})  → broadcast + persist
session_report(model, tokens_in, tokens_out)           → report metrics

MCP Tools — Core Profile (20)

ACP & Identity

ToolDescription
heartbeatSend heartbeat. Call at session start with status=online
get_project_knowledgeGet ALL permanent knowledge (decisions, conventions). ACP Step 2
get_my_rulesGet ALL governance rules (workflow + assignment). ACP Step 3
get_contextGet recent activity + project knowledge. ACP Step 4
get_my_tasksGet assigned tasks. ACP Step 5

Task Management

ToolDescription
list_projectsList workspace projects
list_tasksList tasks with filters (status, priority, assignee, search)
get_taskGet task details with optional comments/artifacts/deps
create_taskCreate a new task
update_taskUpdate task fields
move_taskChange task status using slugs
assign_taskAssign/unassign a task
get_task_contextGet everything about a task in one call

Communication

ToolDescription
add_commentAdd comment to a task (markdown)
publish_eventPublish event + optional memory hint for persistence

Memory

ToolDescription
recallSearch memory by keywords
rememberSave knowledge (UPSERT by key)
forgetDelete a memory entry

Utility

ToolDescription
report_errorReport an error on a task
session_reportReport session metrics (model, tokens, cost)

MCP Tools — Full Profile (adds 25 more)

Additional Task Tools

ToolDescription
get_projectGet project details with statuses and custom fields
create_subtaskCreate subtask under a parent
add_dependencyAdd dependency between tasks
checkout_taskAtomic task lock for multi-agent coordination
release_taskRelease atomic task lock

Comments & Artifacts

ToolDescription
list_commentsList task comments
upload_artifactUpload file/code/log to a task
list_artifactsList task artifacts
get_artifactGet artifact details and download URL

Event Bus

ToolDescription
publish_summaryPublish work summary (convenience wrapper)
subscribe_eventsConfigure webhook delivery for events
poll_tasksLong-poll for new task assignments

Agent & Team

ToolDescription
register_sub_agentRegister a sub-agent
list_sub_agentsList sub-agents (optionally recursive)
get_team_directoryGet workspace team directory
update_agent_profileUpdate agent role, capabilities, profile

Governance & Config

ToolDescription
get_project_rulesGet all project rules
get_assignment_rulesGet assignment rules
get_workflow_rulesGet workflow rules with caller permissions
import_workspace_configImport workspace config from YAML
export_workspace_configExport workspace config as YAML

Recurring Tasks

ToolDescription
create_recurring_taskCreate recurring task schedule
list_recurring_schedulesList recurring schedules
get_recurring_historyGet instance history for a schedule
trigger_recurring_nowTrigger next instance immediately

Architecture

AI Agent (Claude Code / Cursor / Cline / OpenClaw)
    ↕ MCP (stdio or SSE)
EVC Mesh MCP Server (core or full profile)
    ↕ REST API (HTTP)
EVC Mesh API Server
    ↕
PostgreSQL / Redis / NATS / S3

The MCP server is a lightweight proxy — it translates MCP tool calls into REST API requests. No direct database access needed.

Deploy checklist (prod server — no CI)

evc-mesh-mcp has no CD pipeline; deploys are manual. Mandatory order (per CLAUDE-workflow.md §1b Deploy Discipline): migrate (goose up) → binary swap → restart. Never swap the binary before migrations pass.

# 1. Build for the prod target
GOOS=linux GOARCH=amd64 go build -o evc-mesh-mcp .

# 2. Copy binary to prod
scp evc-mesh-mcp root@prod-host:/opt/evc-mesh-mcp/evc-mesh-mcp.new

# 3. On the prod host: run migrations FIRST, then swap binary
ssh root@prod-host

  # STEP 1 — Run evc-mesh DB migrations (mcp server reads the same DB).
  # goose CLI is not installed on the host — use the official docker image.
  # If this exits non-zero, STOP — do NOT swap the binary.
  DB_URL=$(grep ^DATABASE_URL /opt/evc-mesh/.env.prod | cut -d= -f2-)
  docker run --rm --network host \
    -v /opt/evc-mesh/migrations:/migrations \
    ghcr.io/pressly/goose:latest \
    goose -dir /migrations postgres "$DB_URL" up

  # STEP 2 — Swap binary (only after migrations succeed)
  mv /opt/evc-mesh-mcp/evc-mesh-mcp /opt/evc-mesh-mcp/evc-mesh-mcp.bak.$(date +%Y%m%d-%H%M%S)
  mv /opt/evc-mesh-mcp/evc-mesh-mcp.new /opt/evc-mesh-mcp/evc-mesh-mcp

  # STEP 3 — Restart
  sudo systemctl restart evc-mesh-mcp

  # STEP 4 — Smoke test
  curl -sf http://localhost:8081/health || echo "SMOKE FAILED"

evc-mesh-mcp does not run its own migrations — it relies on the evc-mesh API's schema. The goose step above ensures the schema matches before the new binary serves traffic.

Related

License

MIT

Ver o original no GitHub

Bom para