GitHub MCP Server — AutomateHub
Related Servers
Alternatives to GitHub MCP Server — AutomateHub
No user-submitted related servers found.
Related Servers
- AlicenseAqualityBmaintenanceEnables AI agents to operate GitHub through natural language by exposing repository, issue, branch, commit, and pull request operations as validated MCP tools. It connects over stdio to the Antigravity IDE and returns classified, human-readable errors so the agent can act on and explain results directly in the editor.10MIT
- FlicenseNot gradedqualityCmaintenanceEnables AI clients such as Cursor to interact with GitHub repositories and perform issue listing, triage, health checks, prioritization, commenting, weekly digests, and release-note generation through purpose-built MCP tools.-
- AlicenseNot gradedqualityCmaintenanceEnables MCP-compatible AI clients to manage GitHub repositories and issues and trigger repository dispatch events via tools and a REST API.MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to work with multiple GitHub accounts at once, automatically routing each call to the correct token based on an explicit account argument, the repository owner's login, or a configured default. Exposes tools for listing repos, issues, and pull requests plus creating repos, issues, comments, and PRs, while keeping tokens hidden and requiring confirmation before writes.MIT
- AlicenseNot gradedqualityCmaintenanceExposes an MCP endpoint over JSON-RPC 2.0 with seven tools that let agents create and manage repositories, issues, pull requests, labels, and comments, all backed by a shared store so changes are visible across the git, REST v3, and GraphQL surfaces.MIT
- AlicenseAqualityCmaintenanceEnables AI assistants and MCP hosts to inspect GitHub repositories, pull requests, issues, and activity, and to create issues through schema-validated tools, resources, and prompts, with write actions gated by host or CLI confirmation.5MIT
TDQS
Scored across 10 tools
The tools cover clearly distinct resources and actions, and the descriptions include explicit cross-references (e.g. create_issue vs. list_issues vs. add_comment_to_issue) that steer an agent correctly. The one weak spot is create_commit, whose name suggests a commit operation but whose behavior is actually file create/replace, and it could be confused with a broader version-control action.
Nine of ten tools follow a predictable verb_noun pattern (create_*, list_*, close_issue, add_comment_to_issue), with consistent singular/plural handling for create_repository vs. list_repositories. health_check breaks the convention as a noun_noun outlier, and add_comment_to_issue is longer than the rest, but overall readable and predictable.
Ten tools is well within the ideal 3-15 range and each one maps to a concrete GitHub operation. Nothing feels padded or redundant, and the inclusion of a diagnostic health_check is justified for a token-authenticated integration.
The surface covers repository creation/listing, issues (create/list/close/comment), branches, and commits, but there are no pull request tools at all despite PRs being central to GitHub, and no way to read file contents, fetch a single issue's detail, or update/reopen an issue. Agents will hit dead ends as soon as a workflow moves past the narrow issue-and-commit path.