xapi-to 0.1.17 → 0.1.19

package/skills/xapi/guides/tiktok.md

@@ -0,0 +1,322 @@
+# TikTok Guide
+
+Complete guide for TikTok operations via xAPI — user profiles, videos, comments, search, hashtags, music, live rooms, and recommended feed.
+
+> **Dynamic catalog:** These are database-registered third-party APIs under the `tiktok` service. Exact action IDs, HTTP methods, parameters, and response fields can change. Run `search` and `get` before calling; the current schema wins. Examples below reflect one known GET-based version and keep `"method":"GET"` in the input for compatibility.
+
+## Contents
+
+- [User data](#user-data)
+- [Video data](#video-data)
+- [Search](#search)
+- [Hashtags](#hashtags)
+- [Music](#music)
+- [Live](#live)
+- [Feed](#feed)
+- [Common workflows](#common-workflows)
+- [Pagination](#pagination-patterns)
+- [API reference](#api-reference)
+- [Error handling](#error-handling)
+
+## User Data
+
+### Get user profile
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_handler__user__profile \
+  --input '{"method":"GET","params":{"unique_id":"tiktok"}}'
+```
+
+Look up a user by `unique_id` (username), `user_id` (numeric), or `sec_user_id`. At least one must be provided.
+
+Returns `data.data.user` with fields: `uid`, `unique_id`, `nickname`, `signature`, `follower_count`, `following_count`, `total_favorited`, `aweme_count`, `avatar_thumb`, `avatar_medium`, `sec_uid`.
+
+**Important:** The user profile returns `sec_uid`, but other TikTok APIs expect the parameter name `sec_user_id`. Pass the value of `sec_uid` as `sec_user_id` when calling other endpoints.
+
+### Get user_id and sec_user_id by username
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_get__user__id__and__sec__user__id__by__username \
+  --input '{"method":"GET","params":{"username":"tiktok"}}'
+```
+
+**Important:** Many TikTok actions require `sec_user_id` instead of the username. Use this endpoint or `handler_user_profile` to get it first.
+
+### Get user's videos
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__user__post__videos \
+  --input '{"method":"GET","params":{"unique_id":"tiktok","count":10}}'
+```
+
+Can also pass `sec_user_id` instead of `unique_id`. Paginate with `max_cursor` from previous response.
+
+Optional parameters:
+- `count` — items per page
+- `sort_type` — sort type
+- `max_cursor` — pagination cursor
+
+For faster response with simplified data, use the V3 endpoint: `tiktok.api_v1_tiktok_app_v3_fetch__user__post__videos__v3`.
+
+### Get user's liked videos
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__user__like__videos \
+  --input '{"method":"GET","params":{"sec_user_id":"<sec_user_id>","counts":10}}'
+```
+
+Requires `sec_user_id`. Paginate with `max_cursor`. Note: the parameter is `counts` (with 's'), not `count`.
+
+### Get followers / following
+
+```bash
+# Followers
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__user__follower__list \
+  --input '{"method":"GET","params":{"user_id":"107955","count":20}}'
+
+# Following
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__user__following__list \
+  --input '{"method":"GET","params":{"user_id":"107955","count":20}}'
+```
+
+Pass either `user_id` or `sec_user_id`. Paginate with `min_time` and `page_token` from previous response.
+
+### Get similar user recommendations
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__similar__user__recommendations \
+  --input '{"method":"GET","params":{"user_id":"107955"}}'
+```
+
+## Video Data
+
+### Get video by ID
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__one__video__v2 \
+  --input '{"method":"GET","params":{"aweme_id":"<video_id>"}}'
+```
+
+The `aweme_id` is the numeric video ID. Returns full video data including `desc`, `statistics`, `author`, `music`, `video` (with play URLs).
+
+### Get video by share URL
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__one__video__by__share__url__v2 \
+  --input '{"method":"GET","params":{"share_url":"https://www.tiktok.com/@user/video/1234567890"}}'
+```
+
+### Batch get videos
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__multi__video__v2 \
+  --input '{"method":"GET","params":{"aweme_ids":"id1,id2,id3"}}'
+```
+
+### Get video comments
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__video__comments \
+  --input '{"method":"GET","params":{"aweme_id":"<video_id>","count":20,"cursor":0}}'
+```
+
+Paginate with `cursor`.
+
+### Get comment replies
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__video__comment__replies \
+  --input '{"method":"GET","params":{"item_id":"<video_id>","comment_id":"<comment_id>","count":20,"cursor":0}}'
+```
+
+## Search
+
+### General search (comprehensive)
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__general__search__result \
+  --input '{"method":"GET","params":{"keyword":"cat","count":10}}'
+```
+
+Returns mixed results (videos, users, hashtags). Paginate with `offset`.
+
+Optional parameters:
+- `sort_type` — sort type
+- `publish_time` — publish time filter
+
+### Video search
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__video__search__result \
+  --input '{"method":"GET","params":{"keyword":"cooking","count":10}}'
+```
+
+Optional: `region`, `sort_type`, `publish_time`. Paginate with `offset`.
+
+### User search
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__user__search__result \
+  --input '{"method":"GET","params":{"keyword":"chef","count":10}}'
+```
+
+Optional filters: `user_search_follower_count`, `user_search_profile_type`. Paginate with `offset`.
+
+### Music search
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__music__search__result \
+  --input '{"method":"GET","params":{"keyword":"pop","count":10}}'
+```
+
+Optional: `region`, `filter_by`, `sort_type`. Paginate with `offset`.
+
+### Hashtag search
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__hashtag__search__result \
+  --input '{"method":"GET","params":{"keyword":"dance","count":10}}'
+```
+
+Paginate with `offset`.
+
+## Hashtags
+
+### Get hashtag details
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__hashtag__detail \
+  --input '{"method":"GET","params":{"ch_id":"<hashtag_id>"}}'
+```
+
+### Get hashtag video list
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__hashtag__video__list \
+  --input '{"method":"GET","params":{"ch_id":"<hashtag_id>","count":20,"cursor":0}}'
+```
+
+Paginate with `cursor`.
+
+## Music
+
+### Get music details
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__music__detail \
+  --input '{"method":"GET","params":{"music_id":"<music_id>"}}'
+```
+
+### Get videos using a music
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__music__video__list \
+  --input '{"method":"GET","params":{"music_id":"<music_id>","count":20,"cursor":0}}'
+```
+
+Paginate with `cursor`.
+
+## Live
+
+### Get live room info
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__live__room__info \
+  --input '{"method":"GET","params":{"room_id":"<room_id>"}}'
+```
+
+### Check if live room is online
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_check__live__room__online \
+  --input '{"method":"GET","params":{"room_id":"<room_id>"}}'
+```
+
+### Batch check live rooms online
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_check__live__room__online__batch \
+  --input '{"method":"GET","params":{"room_ids":"id1,id2,id3"}}'
+```
+
+### Get live search results
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__live__search__result \
+  --input '{"method":"GET","params":{"keyword":"gaming"}}'
+```
+
+## Feed
+
+### Get home feed (recommended videos)
+
+```bash
+npx xapi-to call tiktok.api_v1_tiktok_app_v3_fetch__home__feed \
+  --input '{"method":"GET","params":{}}'
+```
+
+Returns recommended "For You" page videos. **Note:** This endpoint may return empty results depending on account/region configuration.
+
+## Common Workflows
+
+### Research a TikTok creator
+
+1. Look up user: `tiktok...handler__user__profile` with `unique_id` → get `uid`, `sec_user_id`, follower stats
+2. Get videos: `tiktok...fetch__user__post__videos` → browse recent content
+3. Check engagement: view `statistics` in each video (play count, likes, comments, shares)
+
+### Analyze a TikTok video
+
+1. Get video: `tiktok...fetch__one__video__v2` with `aweme_id` → full video data
+2. Read comments: `tiktok...fetch__video__comments` → top comments
+3. Get replies: `tiktok...fetch__video__comment__replies` → comment threads
+
+### Discover content by hashtag
+
+1. Search hashtags: `tiktok...fetch__hashtag__search__result` with `keyword` → find hashtag IDs
+2. Get details: `tiktok...fetch__hashtag__detail` with `ch_id` → view count, video count
+3. Browse videos: `tiktok...fetch__hashtag__video__list` → videos using this hashtag
+
+## Pagination Patterns
+
+| Pattern | Endpoints | How to use |
+|---------|-----------|------------|
+| `max_cursor` | user videos, liked videos | Pass `max_cursor` from previous response |
+| `cursor` | comments, hashtag videos, music videos | Pass `cursor` from previous response |
+| `offset` | all search endpoints | Increment by `count` each page |
+| `min_time` + `page_token` | follower/following lists | Pass both from previous response |
+
+## API Reference
+
+| API (prefix: `tiktok.api_v1_tiktok_app_v3_`) | Description | Key Params |
+|---|---|---|
+| `handler__user__profile` | Get user profile | `unique_id` or `user_id` or `sec_user_id` |
+| `get__user__id__and__sec__user__id__by__username` | Get IDs by username | `username` |
+| `fetch__user__post__videos` | User's videos | `unique_id` or `sec_user_id`, `count`, `max_cursor` |
+| `fetch__user__post__videos__v3` | User's videos (simplified) | `unique_id` or `sec_user_id`, `count`, `max_cursor` |
+| `fetch__user__like__videos` | User's liked videos | `sec_user_id`, `counts`, `max_cursor` |
+| `fetch__user__follower__list` | User's followers | `user_id` or `sec_user_id`, `count` |
+| `fetch__user__following__list` | User's following | `user_id` or `sec_user_id`, `count` |
+| `fetch__one__video__v2` | Single video | `aweme_id` |
+| `fetch__one__video__by__share__url__v2` | Video by share URL | `share_url` |
+| `fetch__multi__video__v2` | Batch get videos | `aweme_ids` |
+| `fetch__video__comments` | Video comments | `aweme_id`, `count`, `cursor` |
+| `fetch__video__comment__replies` | Comment replies | `item_id`, `comment_id`, `count`, `cursor` |
+| `fetch__general__search__result` | General search | `keyword`, `count`, `offset` |
+| `fetch__video__search__result` | Video search | `keyword`, `count`, `offset` |
+| `fetch__user__search__result` | User search | `keyword`, `count`, `offset` |
+| `fetch__music__search__result` | Music search | `keyword`, `count`, `offset` |
+| `fetch__hashtag__search__result` | Hashtag search | `keyword`, `count`, `offset` |
+| `fetch__hashtag__detail` | Hashtag details | `ch_id` |
+| `fetch__hashtag__video__list` | Hashtag videos | `ch_id`, `count`, `cursor` |
+| `fetch__music__detail` | Music details | `music_id` |
+| `fetch__music__video__list` | Music videos | `music_id`, `count`, `cursor` |
+| `fetch__live__room__info` | Live room info | `room_id` |
+| `check__live__room__online` | Check live online | `room_id` |
+| `fetch__home__feed` | Home feed | — |
+
+## Error Handling
+
+- **Missing sec_user_id** → Use `handler_user_profile` or `get_user_id_and_sec_user_id_by_username` to get it first
+- **Empty results** → Check that `aweme_id`, `ch_id`, or `music_id` is valid
+- **Pagination exhausted** → When `has_more` is `0` or `false`, no more pages available

package/skills/xapi/guides/twitter.md

@@ -0,0 +1,276 @@
+# Twitter / X Guide
+
+Complete guide for Twitter operations via xAPI — reading data, downloading tweet videos, posting tweets, replying, and OAuth setup.
+
+> **Upstream provider:** All `twitter.*` capabilities accept an optional `provider` — `"x"` (fapi.uk, the default) or `"twitter"` (legacy upstream). The response is normalized to an identical structure regardless of provider, so you normally omit it. Pass `"provider":"twitter"` only to force the legacy upstream, e.g. `--input '{"screen_name":"elonmusk","provider":"twitter"}'`.
+
+## Contents
+
+- [Read Twitter data](#reading-twitter-data-no-oauth-needed)
+- [Download a tweet video](#download-a-tweet-video)
+- [Post and mutate with OAuth](#posting-tweets-oauth-required)
+- [Common workflows](#common-workflows)
+- [Pagination reference](#pagination-reference)
+- [API reference](#api-reference)
+- [Error handling](#error-handling)
+
+## Reading Twitter Data (no OAuth needed)
+
+### Look up a user by @handle
+
+```bash
+npx xapi-to call twitter.user_by_screen_name --input '{"screen_name":"elonmusk"}'
+```
+
+Returns `rest_id` (numeric user ID), `name`, `screen_name`, `followers_count`, `statuses_count`, etc.
+
+**Important:** Most Twitter actions require the numeric `user_id` (called `rest_id` in the response), not the @handle. Always look up the user first to get the ID.
+
+### Get a user's recent tweets
+
+```bash
+npx xapi-to call twitter.user_tweets --input '{"user_id":"44196397"}'
+```
+
+Each tweet includes: `id`, `full_text`, `created_at`, `favorite_count`, `retweet_count`, `reply_count`, `views_count`, `media`, `author`, and `quoted_tweet` if applicable.
+
+Paginate with the previous response's `data.cursors.bottom`:
+
+```bash
+npx xapi-to call twitter.user_tweets \
+  --input '{"user_id":"44196397","cursor":"<cursors.bottom>"}'
+```
+
+### Get a user's tweets and replies
+
+```bash
+npx xapi-to call twitter.user_tweets_and_replies --input '{"user_id":"44196397"}'
+```
+
+Similar to `twitter.user_tweets`, but the timeline also includes the user's replies to other tweets and conversation threads they participate in. Each item includes the same fields: `id`, `full_text`, `created_at`, `favorite_count`, `retweet_count`, `reply_count`, `views_count`, `media`, `author`, and `quoted_tweet` if applicable.
+
+**When to choose which:**
+
+- `twitter.user_tweets` — only the user's own posts. Use for content/timeline of original tweets.
+- `twitter.user_tweets_and_replies` — posts plus replies and conversation participants. Use to monitor a user's reply activity or full timeline.
+
+**Filter tip:** because conversation entries can contain tweets from other authors, filter by `author.id === user_id` if you only want the monitored user's content.
+
+Pagination uses the same `cursor` → `data.cursors.bottom` pattern as `twitter.user_tweets`.
+
+For `twitter.user_tweets` and `twitter.user_tweets_and_replies`, set `"cache":true` to use the fast cache when available and skip the normal upstream call. Cached results can be less fresh; keep the default `false` when freshness matters.
+
+### Get a specific tweet and its replies
+
+```bash
+npx xapi-to call twitter.tweet_detail --input '{"tweet_id":"2035526376468394305"}'
+```
+
+To load more replies, pass the previous response's `data.cursors.bottom` as `cursor`:
+
+```bash
+npx xapi-to call twitter.tweet_detail \
+  --input '{"tweet_id":"2035526376468394305","cursor":"<cursors.bottom>"}'
+```
+
+Paginated pages commonly contain more `replies` but no main tweet, so `data.tweet` may be `null` when `cursor` is present.
+
+### Download a tweet video
+
+Use the bundled downloader with either a complete status URL or its numeric tweet ID. Resolve `<xapi-skill-directory>` to the directory containing this guide's parent `SKILL.md`:
+
+```bash
+bash <xapi-skill-directory>/scripts/download_tweet_videos.sh \
+  'https://x.com/NousResearch/status/2084325600643445095'
+```
+
+Pass an optional existing output directory as the second argument; otherwise files are written to the current directory. A single video is named `tweet-<tweet-id>.mp4`; multiple videos are numbered `tweet-<tweet-id>-1.mp4`, `-2.mp4`, and so on. The script includes videos in the main, quoted, and retweeted tweets, deduplicates identical media URLs, and refuses to overwrite any existing output.
+
+The script verifies that the deployed `twitter.tweet_detail` schema exposes `video_url` before making the paid call. It omits `provider` to retain automatic failover from the default `x` upstream to the legacy `twitter` upstream on transient failures. The response's `data.provider` identifies the upstream that served it; explicitly setting `provider` would pin that upstream and disable failover.
+
+Each download is written to a same-directory temporary file, required to have MIME type `video/mp4`, optionally checked with `ffprobe`, and only renamed to its final path after every media file passes. Failures clean up temporary files, so a partial transfer is never published as the final MP4.
+
+`data.tweet.media[].url` and `preview_url` are preview images; do not download them as MP4. The optional `video_info.variants` retains the original MP4 and HLS variants when a specific rendition is needed. `call --output` is not appropriate because the xAPI action returns JSON metadata rather than video bytes; the bundled script downloads each resolved `video_url` with `curl`.
+
+### Read an X Article (long-form post)
+
+Some tweets are long-form **X Articles**. `twitter.tweet_detail` detects them and returns the full normalized article directly in `data.tweet.article`; no raw GraphQL call or `fieldToggles` parameter is needed.
+
+```bash
+npx xapi-to call twitter.tweet_detail --input '{"tweet_id":"<tweet_id>"}'
+```
+
+The normalized `article` object contains:
+
+- `id`, `title`, and `preview_text`
+- `text` — full plain text
+- `markdown` — full text with headings, lists, quotes, code blocks, and inline links preserved
+- `cover_image` — URL and optional dimensions
+- `links` — deduplicated external links in appearance order
+- `first_published_at` and `modified_at` — ISO 8601 timestamps when available
+
+Use the **tweet ID** from the share URL (`x.com/<user>/status/<tweet_id>` or `x.com/<user>/article/<tweet_id>`), not an internal article ID.
+
+### Search tweets
+
+```bash
+npx xapi-to call twitter.search --input '{"raw_query":"AI agents","count":20}'
+```
+
+For `provider: "x"` (the default), structured advanced-search filters are also available:
+
+```bash
+npx xapi-to call twitter.search --input '{
+  "raw_query":"AI",
+  "from":"OpenAI",
+  "mentioning":"AnthropicAI",
+  "phrase":"AI agents",
+  "since":"2026-08-01",
+  "until":"2026-08-05",
+  "min_likes":100,
+  "min_replies":10,
+  "min_retweets":20,
+  "count":20
+}'
+```
+
+Supported structured filters: `from`, `to`, `mentioning`, `phrase`, `any`, `none`, `tag`, `since`, `until`, `min_replies`, `min_likes`, `min_retweets`, and `count`. `sort_by` accepts `Top`, `Latest` (default), `People`, `Photos`, or `Videos`; the default `x` provider maps both media-specific values to its combined Media search. Dates use `YYYY-MM-DD`; `until` is exclusive. Paginate with `cursor` from `data.cursor_bottom`.
+
+### Get user's media posts
+
+```bash
+npx xapi-to call twitter.user_media --input '{"user_id":"44196397"}'
+```
+
+Paginate with `cursor` from the previous response's `data.cursor_bottom`. Both the default `x` provider and legacy `twitter` provider are supported.
+
+### Get followers / following
+
+```bash
+npx xapi-to call twitter.followers --input '{"user_id":"44196397"}'
+npx xapi-to call twitter.following --input '{"user_id":"44196397"}'
+```
+
+### Get retweeters
+
+```bash
+npx xapi-to call twitter.retweeters --input '{"tweet_id":"1234567890"}'
+```
+
+## Posting Tweets (OAuth required)
+
+Posting, replying, quoting, liking, retweeting, and deleting require OAuth. A saved binding provides technical authorization, not standing user consent: confirm the current content and target before a write, and obtain explicit confirmation before destructive or bulk actions.
+
+### Step 1: Bind Twitter OAuth
+
+```bash
+npx xapi-to oauth bind --provider twitter
+```
+
+This opens a browser for the user to authorize. After authorization, the binding is saved to the API key.
+
+Verify:
+
+```bash
+npx xapi-to oauth status
+```
+
+Should show `tweet.write` in scopes.
+
+### Step 2: Post a tweet
+
+```bash
+npx xapi-to call x-official.2_tweets --method POST \
+  --input '{"body":{"text":"Hello from my AI agent!"}}'
+```
+
+**Character limit:** 280 characters (140 CJK characters). Each CJK character counts as 2.
+
+### Reply to a tweet
+
+```bash
+npx xapi-to call x-official.2_tweets --method POST \
+  --input '{"body":{"text":"Great point!","reply":{"in_reply_to_tweet_id":"2035526376468394305"}}}'
+```
+
+### Quote tweet
+
+```bash
+npx xapi-to call x-official.2_tweets --method POST \
+  --input '{"body":{"text":"Worth reading 👇","quote_tweet_id":"2035526376468394305"}}'
+```
+
+### Delete a tweet
+
+```bash
+npx xapi-to call x-official.2_tweets_id --method DELETE \
+  --input '{"pathParams":{"id":"2036012345678901234"}}'
+```
+
+### Like a tweet
+
+```bash
+npx xapi-to call x-official.2_users_id_likes --method POST \
+  --input '{"pathParams":{"id":"<your_user_id>"},"body":{"tweet_id":"2035526376468394305"}}'
+```
+
+### Retweet
+
+```bash
+npx xapi-to call x-official.2_users_id_retweets --method POST \
+  --input '{"pathParams":{"id":"<your_user_id>"},"body":{"tweet_id":"2035526376468394305"}}'
+```
+
+## Common Workflows
+
+### Research and tweet
+
+1. Search the web: `web.search.realtime` → get latest news
+2. Summarize: `ai.text.summarize` → create a concise summary
+3. Post: `x-official.2_tweets` POST → tweet the summary
+
+### Monitor and reply
+
+1. Get user activity: `twitter.user_tweets` (originals only) or `twitter.user_tweets_and_replies` (includes replies) → check latest posts
+2. Get tweet detail: `twitter.tweet_detail` → read the thread
+3. Reply: `x-official.2_tweets` POST with `reply` → respond
+
+## Pagination Reference
+
+| Capability | Next cursor field | Next request input |
+|---|---|---|
+| `twitter.user_tweets` | `data.cursors.bottom` | `cursor` |
+| `twitter.user_tweets_and_replies` | `data.cursors.bottom` | `cursor` |
+| `twitter.tweet_detail` replies | `data.cursors.bottom` | `cursor` |
+| `twitter.user_media` | `data.cursor_bottom` | `cursor` |
+| `twitter.search` | `data.cursor_bottom` | `cursor` |
+| `twitter.followers` | `data.cursor_bottom` | `cursor` |
+| `twitter.following` | `data.cursor_bottom` | `cursor` |
+| `twitter.retweeters` | `data.cursor_bottom` | `cursor` |
+
+Omit `cursor` for the first page. Stop when the relevant bottom cursor is absent or empty.
+
+## API Reference
+
+| API | Method | Description |
+|-----------|--------|-------------|
+| `twitter.user_by_screen_name` | — | Look up user by @handle |
+| `twitter.user_tweets` | — | Get and paginate user's recent tweets |
+| `twitter.user_tweets_and_replies` | — | Get and paginate user's tweets and replies |
+| `twitter.user_media` | — | Get and paginate user's media posts |
+| `twitter.tweet_detail` | — | Get tweet, full X Article content, and paginated replies |
+| `twitter.search` | — | Search tweets with cursor and advanced filters |
+| `twitter.followers` | — | Get user's followers |
+| `twitter.following` | — | Get user's following |
+| `twitter.retweeters` | — | Get tweet retweeters |
+| `x-official.2_tweets` | POST | Post a tweet |
+| `x-official.2_tweets_id` | DELETE | Delete a tweet |
+| `x-official.2_users_id_likes` | POST | Like a tweet |
+| `x-official.2_users_id_retweets` | POST | Retweet |
+
+## Error Handling
+
+- **OAuth Required** → Run `npx xapi-to oauth bind --provider twitter`
+- **403 Forbidden** → Twitter account may have restrictions; check account status
+- **Tweet too long** → Shorten to 280 chars (140 CJK)
+- **User not found** → Check the screen_name spelling