F
licenseA
qualityD
maintenanceAn MCP server that orchestrates parallel Claude agent sessions by decomposing natural-language tasks into contract-bound tenants with structured outputs and validation, all running over stdio.
5
-
swarm-mcp
by stiege
Related Servers
Alternatives to swarm-mcp
No user-submitted related servers found.
Related Servers
- AlicenseNot gradedqualityAmaintenanceRun dynamic, multi-agent workflow scripts — agent(), parallel(), pipeline() — over real coding agents (Claude Code and OpenAI Codex), with deterministic journaling, resume, token budgets, and git-worktree isolation.7Apache 2.0
- AlicenseNot gradedqualityCmaintenanceCoordinates up to 64 parallel expert agents for complex multi-agent orchestration and intelligent task routing within Claude Code. It enables users to analyze requests, manage hierarchical execution plans, and monitor active orchestration sessions.MIT
- AlicenseCqualityAmaintenanceEnables multiple CLI-based AI agents to collaborate as a coordinated team through shared task queues, shared memory, and a message bus, with DAG orchestration, rate-limit avoidance, parallel dispatching, and long-task management.632MIT
- FlicenseBqualityNot gradedmaintenanceBridges Claude Code to a Cloud Orchestrator API, providing access to multi-AI consensus, web search, code execution sandboxes, long-term memory, knowledge graphs, deployment management, and 20+ integrated AI and developer tools.28-
- AlicenseNot gradedqualityBmaintenanceEnables concurrent multi-agent orchestration, running many AI tasks in parallel over a local OpenAI-compatible router or the Antigravity CLI, with tools for model discovery, health checks, logging, and batch execution.GPL 3.0
TDQS
A3.8/5.0
Scored across 33 tools
Disambiguation4/5
Most tools have distinct purposes, but there are some overlaps that could cause confusion, such as 'chain' vs 'pipeline' and 'map' vs 'map_reduce'. Descriptions help differentiate them, but an agent might struggle to pick the right one in some cases.
Naming Consistency3/5
Names follow snake_case, but the pattern is inconsistent: some are verbs (run, validate, wrap), some are nouns (beam, chain, map), and some are abbreviations (par). This lack of a unified naming convention reduces predictability.
Tool Count3/5
With 33 tools, the server is on the larger side. While each tool serves a specific purpose in a complex orchestration system, the count exceeds the typical 3-15 range, making it feel heavy but not extreme.
Completeness4/5
The tool set covers execution, evaluation, state management, type system, pipeline management, and governance comprehensively. Minor gaps exist, such as lack of a general run listing tool, but overall the surface is broad and well-rounded.
Maintenance
ActivityInactive
ResponsivenessNo issues