A
licenseA
qualityA
maintenanceEnables AI assistants to delegate specific tasks to specialized sub-agents (e.g., test-writer, code-reviewer). Supports both Cursor and Claude Code with custom agent definitions.
1
893
96
MIT
SubagentBridge
by rzqllh
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@SubagentBridgeSpawn a background subagent to refactor the auth module and report back."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
SubagentBridge
An MCP server that lets a primary agent inside Antigravity IDE delegate coding tasks to background subagents across multiple model backends — with automatic crash recovery, guaranteed process cleanup, and a real Human-in-the-Loop approval gate for destructive actions.
Architecturally inspired by antigravity-mcp
by Inferno-Aditya (MIT License). No code from that project is reused directly —
see THIRD_PARTY_NOTICES.md for full attribution.
Why
A review of antigravity-mcp's source found it didn't fully deliver on its own
docs: timed-out subagents were left running, failed runs weren't retried, a
workspace-boundary check could silently fall through, and the advertised
"Human-in-the-Loop" safety gate didn't actually exist in code. SubagentBridge
fixes all four and adds first-class support for GPT-5.6 (Sol/Luna) and
DeepSeek V4 Flash as additional runners. Full rationale in docs/PRD.md.
Related MCP server: any-model-plugin
Status
🚧 Planning stage — see docs/ for PRD, architecture, weakness tracker, API
reference, runner spec, security notes, and test plan. Implementation not yet
started.
Planned Tools
Tool | Purpose |
| Start a new background subagent session |
| Send a follow-up instruction to a session |
| Block until a session finishes or times out (kills the process on timeout) |
| Inspect running/finished sessions |
| Token/cost accounting per session |
| Direct surgical find-replace patch, workspace-bounded |
| Force-stop a session |
| Resolve a pending HITL approval request |
See docs/API_REFERENCE.md for full signatures.
Supported Runners
agy— Antigravity IDE's own agentclaude— Claude Codegpt_sol— GPT-5.6 Sol/Luna (planned)deepseek— DeepSeek V4 Flash (planned)
Setup (planned)
pip install -r requirements.txtAdd to Antigravity's mcp_config.json — see mcp_config.example.json.
License
MIT — see LICENSE. Third-party attribution in THIRD_PARTY_NOTICES.md.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
No tool schema history has been recorded yet.
This server cannot be installed
Maintenance
ActivityMaintained
ResponsivenessNo issues
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Connectors
- ParleyOAuthdev.weldra
Coordination hub for AI coding agents: message teammates, ask humans, audit every event.
Shared control plane for AI coding agents — tasks, memory, decisions, file locks. 12 tools.
- projectsOAuthcloud.tri2b
Task tracking built for coding agents. Work is leased, so two agents never take the same SubTask.
- AgentdaOAuthcom.myagentda
Agent-native task management: your AI agent is the interface. Delegate to anyone by email.
Related MCP Servers
- AlicenseNot gradedqualityAmaintenanceEnables AI agents to delegate tasks, run adversarial reviews, and manage background jobs across multiple models and providers via anymodel_* tools.Apache 2.0
- AlicenseAqualityAmaintenanceEnables delegating asynchronous coding tasks and DAG workflows to local Oh My Pi (OMP) CLI sub-agents, with topological orchestration, path isolation, and supervised resumption.93421MIT
- AlicenseAqualityBmaintenanceEnables AI coding agents like Claude Code or Codex to delegate tasks to a DeepSeek Harness subagent with its own context window, providing tools for task delegation, result waiting, continuation, and supervision with sandboxed execution.6MIT
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/rzqllh/SubagentBridge'
If you have feedback or need assistance with the MCP directory API, please join our Discord server