A
licenseA
qualityF
maintenanceMCP server for running external coding agents as background tasks inside Claude Code. Supports multiple backends including Codex, Grok, GLM, DeepSeek, and more.
7
MIT
lightbulb-mcp
by RPasquale
Related Servers
Alternatives to lightbulb-mcp
No user-submitted related servers found.
Related Servers
- AlicenseAqualityDmaintenanceMCP server that bridges OpenAI's Agents SDK with Claude Code, enabling web search, file search, and computer use capabilities directly in your development environment.29 npm1MIT
- AlicenseNot gradedqualityBmaintenanceMCP server that lets Claude Code or any MCP client delegate coding tasks to the local Cursor Agent CLI, with persistent per-workspace sessions and tools for diffing and pushing changes.12 npmMIT
- FlicenseNot gradedqualityDmaintenanceAn MCP server wrapping the Cursor CLI agent, enabling Claude Code and other MCP clients to delegate tasks to Cursor's AI agent for file writing, bash commands, and codebase queries.-
- AlicenseNot gradedqualityDmaintenanceMCP server that enables Claude Code to communicate with other Claude Code agents over HTTP, allowing users to ask questions about remote codebases or delegate coding tasks.MIT
- AlicenseAqualityBmaintenanceCode intelligence MCP server for Claude Code providing multi-project code graph, semantic search, session history, knowledge base, and web search.153MIT
TDQS
C2.1/5.0
Scored across 1677 tools
Disambiguation1/5
The same underlying operations are exposed through multiple overlapping surfaces (e.g., salesforce_bulk_ingest_close vs connector_salesforce_bulk_ingest_close, quickbooks_company_info vs quickbooks_get_company_info, search_documents vs document_intelligence_search_documents), making it impossible to reliably tell tools apart. Generic lifecycle verbs like run, execute, start, propose, and finalize recur across dozens of similar workflow and company_* tools with little distinguishing structure.
Naming Consistency2/5
The set is mostly snake_case but mixes domain-prefixed dispatcher names (finance_chat, coding_run_tests) with bare platform tools (list_workflows, invoke_tool), and includes many non-verb names like company_chaos, workspace_bundle, and page_builder_workspace_automation. Verb style is inconsistent across lifecycle tools, with get/list/run/trigger/start/propose/execute all used for similar operations.
Tool Count1/5
With 1,677 tools, this is an extreme count that far exceeds any reasonable MCP surface. It reads as a full platform export rather than a curated server, and the sheer volume makes agent tool selection and context management impractical.
Completeness3/5
Domain coverage is remarkably broad across finance, CRM, legal, HR, IT ops, commerce, content, and document intelligence, and most CRUD cycles exist somewhere. However, the surface is fragmented across duplicated layers with notable dead ends, including artifact tools with no update/delete, many preview-only or fail-closed operations, and a training-admission flow that always returns 503.
Maintenance
ActivityMaintained
ResponsivenessNo issues