A
licenseA
qualityC
maintenanceEnables OpenAI Codex to delegate tasks to Google Antigravity CLI, running them headlessly and polling for results.
4
MIT
Codex Antigravity Subagent MCP
Related Servers
Alternatives to Codex Antigravity Subagent MCP
No user-submitted related servers found.
Related Servers
- AlicenseNot gradedqualityAmaintenanceEnables Codex users to delegate engineering, refactoring, auditing, and test execution tasks to Google Antigravity CLI as a persistent autonomous subagent, with tools for managing sessions and running prompts.MIT
- AlicenseBqualityAmaintenanceAn MCP bridge that lets Codex delegate long-running agent work to the Antigravity CLI, providing observable and resumable tool-based execution with project scoping.1473 PyPI12MIT
- AlicenseNot gradedqualityCmaintenanceEnables Claude Desktop to execute tasks via the Google Antigravity CLI (agy), exposing its models, skills, plugins, and background workers as native tools.9MIT
- AlicenseAqualityBmaintenanceEnables Codex to delegate bounded code analysis, review, isolated edits, multimodal research, and native image generation to supervised Antigravity workers, with queued execution, retries, and inspectable patches.232MIT
- AlicenseAqualityCmaintenanceEnables AI harnesses to delegate code analysis, modification, testing, and long-running tasks to the locally installed Antigravity CLI via stdio MCP, with job status tracking and conversation continuity.3MIT
TDQS
A4/5.0
Scored across 8 tools
Disambiguation4/5
Most tools target distinct actions: starting background tasks, starting synchronous tasks, querying, canceling, finishing, interacting, checking status, and opening the dashboard. The only mild ambiguity is between start/delegate and cancel/finish, but the descriptions clearly separate async vs sync and graceful vs hard termination.
Naming Consistency4/5
The naming convention is largely consistent snake_case with verb_object patterns like start_gemini_task, get_gemini_task, and cancel_gemini_task. Minor deviations are delegate_to_gemini and antigravity_status, but they remain readable and predictable.
Tool Count5/5
Eight tools is well-scoped for a task/subagent management server. Each tool covers a meaningful lifecycle or support function without unnecessary duplication or bloat.
Completeness4/5
The core lifecycle is covered: start sync/async, query status/results, cancel, finish, interact, and verify environment. A minor gap is the lack of a way to list all jobs or past tasks without knowing a job_id, though the dashboard partially compensates.
Maintenance
ActivityMaintained
ResponsivenessNo issues