YouTube video/channel search, suggestions, details, comments/replies, Shorts, transcripts, paging.
Scored across 12 tools
Most tools map to distinct resource+action pairs, and the descriptions actively steer the agent (e.g. comments vs. comment replies via reply_token, search vs. detail via link presence). The only real overlap is youtube_submit_video_speech_text_by_url vs _by_video_id, which perform the same operation with different input identifiers, and the get_job tool that belongs to that same workflow.
Nearly all tools follow a predictable snake_case `youtube_<verb>_<noun>` pattern with clear suffixes (_by_url, _by_video_id, _job). The single socialdatax_get_points_balance tool breaks the prefix convention, but it is clearly a different sub-domain (billing), so the deviation is minor and readable.
Twelve tools is well within the ideal 3-15 range and each tool covers a distinct capability (channel lookup, video listing, search, comments, replies, transcript submission/status, balance). No tool feels redundant padding.
The surface covers the core YouTube data workflows end-to-end: channel info, channel videos, video detail, comments, replies, three search variants, and an async transcript submit/status lifecycle. Minor gaps exist around playlists and channel-level transcript/other list endpoints, but agents can work around these.
根据 YouTube 频道主页链接获取频道资料。
Inputs: channel_url (required). Returns: bio, name, points, country, +9 more
根据 YouTube 频道主页链接获取频道发布视频或 Shorts 列表。
Inputs: channel_url (required), page_token, video_type. Returns: items, points, next_page_token
根据 reply_token 获取 YouTube 评论回复；用户已提供有效 reply_token 时直接使用；已有 YouTube 视频链接但缺少 reply_token 时，调用 youtube_get_video_comments_by_url 获取一级评论；缺少视频定位信息时向用户索取；不要传一级评论或回复项的 comment_id；支持 page_token 翻页。
Inputs: reply_token (required), page_token. Returns: items, points, next_page_token
根据 YouTube 视频链接获取公开视频一级评论。
Inputs: video_url (required), sort_type, page_token. Returns: items, points, comment_count, next_page_token
根据 YouTube 视频链接获取公开视频详情。
Inputs: video_url (required). Returns: title, author, points, video_id, +9 more
根据用户提供的有效 job_id，或 submit 工具返回的 job_id 查询 YouTube 视频口播转文字任务状态；每次最多等待 240 秒，不触发重处理，也不要重复提交任务。
Inputs: job_id (required). Returns: error, job_id, status, message, +9 more
按关键词搜索 YouTube 频道或创作者，返回频道摘要，支持 page_token 翻页；已有频道主页链接时读取频道资料，搜索视频时使用视频搜索。
Inputs: keyword (required), page_token. Returns: items, points, next_page_token
获取 YouTube 搜索框联想词，不返回视频或频道详情，不支持分页；可将 text 用于视频搜索，已有明确搜索词时可直接搜索视频；已有视频或频道链接时使用对应详情工具。
Inputs: keyword (required). Returns: items, points
按搜索词搜索 YouTube 公开视频，只返回视频结果。用户需要按搜索词查找视频时使用；已有视频、Shorts 或 youtu.be 链接时使用视频详情或评论工具；已有频道主页链接时使用频道资料或频道发布视频工具；不支持播放列表链接作为搜索输入；支持筛选和 page_token 翻页。
Inputs: keyword (required), sort_type, page_token, video_type, +2 more. Returns: items, points, next_page_token
提交 YouTube 视频口播转文字任务；提交后最多等待 240 秒，未完成时返回 job_id 和下一步查询动作。
Inputs: video_url (required). Returns: error, job_id, status, message, +9 more
根据 YouTube video_id 提交口播转文字任务；提交后最多等待 240 秒，未完成时返回 job_id 和下一步查询动作。
Inputs: video_id (required). Returns: error, job_id, status, message, +9 more
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
YouTube search: find videos, fetch details and comments, pull full transcripts.
YouTube transcripts/subtitles, comments + replies, Shorts and video/channel data.
YouTube search, whole channels, video stats, comments and full transcripts with timestamps.
YouTube transcripts, comments, video details, heatmaps, search, channels and playlists.
Add one secure layer between your agents and this server.
socialdatax_get_points_balanceARead-only Inspect
查询当前 API Key 账户的 SocialDataX 积分余额、剩余积分或点数 / remaining points balance。需要充值时请访问 https://socialdatax.com/dashboard/billing。
Inputs: no named parameters. Returns: object