bluesky-mcp
Related Servers
Alternatives to bluesky-mcp
No user-submitted related servers found.
Related Servers
- AlicenseAqualityBmaintenanceEnables AI agents to post, reply, run threads, read timelines, search, manage follows, and analyze engagement on Bluesky, with setup requiring only a handle and app password.41MIT
- AlicenseNot gradedqualityBmaintenanceEnables interaction with Bluesky social network through the AT Protocol, including searching posts, fetching profiles, browsing feeds, and retrieving threads and follower data.21 npmMIT
- AlicenseAqualityDmaintenanceEnables search, reading, and posting to Bluesky from any MCP client; features 11 tools (5 read, 6 write) with gated writes requiring explicit confirmation to prevent accidental publishing.1161 npmMIT
- FlicenseNot gradedqualityCmaintenanceEnables AI agents to interact with Bluesky social network through optimized, token-efficient tools for posting, replying, searching, and managing social relationships. Features Windows-compatible implementation with context cleanup to minimize LLM token consumption.7-
- AlicenseAqualityCmaintenanceEnables LLMs to interact with the AT Protocol ecosystem, including Bluesky, through natural language. Supports public data access without authentication and full write operations with authentication.5146 npm9MIT
- AlicenseCqualityBmaintenanceEnables interacting with Bluesky/AT Protocol through MCP tools for posting, replying, media, timelines, and a human-approved outbox system.4MIT
TDQS
Scored across 45 tools
Each tool targets a distinct resource-action pair: read vs. write tools are clearly separated (get_post_stats vs. rank_posts vs. get_engagement_summary cover single-post, ranked-set, and aggregate views respectively), and paired operations (like/unlike, follow/unfollow, block/unblock) are unambiguous. Descriptions actively steer selection, e.g. 'the one-post version of rank_posts'.
The set is dominated by a predictable snake_case verb_noun pattern (get_feed, create_post, mute_account, search_actors). A few deviations like whoami, get_unread_count, and mark_notifications_seen break the verb_noun convention slightly but remain readable and non-ambiguous.
At 45 tools this is well past the heavy threshold for a single server surface, forcing an agent to sift through many near-neighbors before acting. The breadth is largely earned by the domain, but the count is still excessive for practical selection.
Coverage is unusually thorough: post lifecycle (create/thread/delete), interactions, social graph, feeds, lists, notifications, and a full engagement-analytics suite. Minor gaps remain—no direct-message/chat tools, no list creation, and no bookmark operations—but core workflows have no dead ends.