SkillAgentSearch skills...

lark-im

飞书即时通讯:收发消息和管理群聊。发送和回复消息、搜索聊天记录、管理群聊成员、上传下载图片和文件、管理表情回复、发送应用内/短信/电话加急、发送和处理交互卡片(Interactive Card)、监听卡片按钮回调(card.action.trigger)。当用户需要发消息、查看或搜索聊天记录、下载聊天中的文件、查看群成员、搜索群、创建群聊或话题群、管理标记数据、管理 Feed 置顶(添加/移除/查询置顶会话)、管理标签数据、处理卡片回调时使用。

Install / Use

npx skills add larksuite/cli --skill lark-im

Installs into whichever agent you are using.

About this skill
📄

SKILL.md

Installable skill definition

Quality Score

96/100

Supported Platforms

Universal

Our assessment of lark-im

lark-im scores 96/100 on our quality scale, 34th of 319 Customer Support skills we index (top 11%).

Its SKILL.md is 23 KB long, well organised into 32 sections with 2 code examples: a thorough specification that gives an agent plenty to work with.

With 17,444 GitHub stars, it is one of the more widely adopted skills in the catalogue.

Substance
30/30
Structure
18/20
Description
15/15
Adoption
18/20
Freshness
15/15

Maintenance, license and trust

  • The repository was last updated about 2 months ago, so lark-im is actively maintained.
  • Our last check on 2026-11-15 found the source still online.
  • It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
  • Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.

Safety scan

No issues found

Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.

AI review by kimi-k2.7-code on 2026-09-28. Automated pattern scan on 2026-09-28. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.

lark-im compared with similar skills

All 4 of these similar skills score higher than lark-im; compare them before choosing.

SkillScoreStarsUpdatedFormat
lark-im (this skill)by larksuite9617.4k53d agoSKILL.md
taste-skillby Leonxlnx100116.8k8d agoCLAUDE.md
LocalAIby mudler10050.4ktodayMCP Server
x402by browser-use100121.0k3d agoSKILL.md
algorithmic-artby anthropics100177.9k55d agoSKILL.md

Frequently asked questions

How do I install lark-im?
Run npx skills add larksuite/cli --skill lark-im. The install tabs above show the steps for each supported agent.
Which AI agents does lark-im work with?
It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
Is lark-im safe to use?
Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It is MIT-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
Is lark-im still maintained?
The repository was last updated about 2 months ago, so lark-im is actively maintained.

name: lark-im version: 1.0.0 description: "飞书即时通讯:收发消息和管理群聊。发送和回复消息、搜索聊天记录、管理群聊成员、上传下载图片和文件、管理表情回复、发送应用内/短信/电话加急、发送和处理交互卡片(Interactive Card)、监听卡片按钮回调(card.action.trigger)。当用户需要发消息、查看或搜索聊天记录、下载聊天中的文件、查看群成员、搜索群、创建群聊或话题群、管理标记数据、管理 Feed 置顶(添加/移除/查询置顶会话)、管理标签数据、处理卡片回调时使用。" metadata: requires: bins: ["lark-cli"] cliHelp: "lark-cli im --help"

im (v1)

CRITICAL — 开始前 MUST 先用 Read 工具读取 ../lark-shared/SKILL.md,其中包含认证、权限处理

Core Concepts

  • Message: A single message in a chat, identified by message_id (om_xxx). Supports types: text, post, image, file, audio, video, sticker, interactive (card), share_chat, share_user, merge_forward, etc.
  • Chat: A group chat or P2P conversation, identified by chat_id (oc_xxx).
  • Thread: A reply thread under a message, identified by thread_id (om_xxx or omt_xxx).
  • Reaction: An emoji reaction on a message.
  • Flag: A bookmark on a message or thread.
  • Feed Shortcut: A chat pinned to the current user's feed sidebar, identified by feed_card_id (an oc_xxx open_chat_id for CHAT type).
  • Feed Group: A tag that groups feed cards in the feed list, identified by feed_group_id (ofg_xxx). Members are feed cards, each identified by feed_id + feed_type. Two types: normal (members managed explicitly) and rule (members auto-derived from rules).

Resource Relationships

Chat (oc_xxx)
├── Message (om_xxx)
│   ├── Thread (reply thread)
│   ├── Reaction (emoji)
│   └── Resource (image / file / video / audio)
└── Member (user / bot)

Important Notes

AppLink and Share Links

Prefer CLI-returned links: use chat_app_link to open joined conversations, message_app_link to open messages, and share_link to invite others to groups. If manually building a joined-conversation AppLink, use https://<applink_host>/client/chat/open?openChatId=<oc_xxx>, never chatId=<oc_xxx> or lark://...chat_id=<oc_xxx>.

Identity and Token Mapping

  • --as user means user identity and uses user_access_token. Calls run as the authorized end user, so permissions depend on both the app scopes and that user's own access to the target chat/message/resource.
  • --as bot means bot identity and uses tenant_access_token. Calls run as the app bot, so behavior depends on the bot's membership, app visibility, availability range, and bot-specific scopes.
  • If an IM API says it supports both user and bot, the token type changes who the operator is. The same API can succeed with one identity and fail with the other because owner/admin status, chat membership, tenant boundary, or app availability are checked against the current caller.

Sender Name Resolution

When fetching messages (+chat-messages-list, +threads-messages-list, +messages-mget, +messages-search), the CLI shows a display name for both user and bot senders:

  • Server-provided name: the read APIs return sender_name (plus the full-i18n sender_i18n_names map) on each message sender; the CLI surfaces it as the sender's name for users and bots alike. No name lookup and no extra permission are needed — no contact scope and no application:bot.basic_info:read.
  • Fallback to id: when the server does not provide a name, the sender is shown by its id and the command still exits 0. There is no contact-directory fallback.

The raw sender_name is not duplicated in output (its value is in name); the full sender_i18n_names map (all locales) is preserved for consumers that need a specific language, alongside an optional open_bot_id (ou_) for bot senders aligned with the message-receive event channel. System messages (msg_type: system) have no sender name — that is normal, not an error.

Default message enrichment (reactions / update_time)

The four message-pulling shortcuts (+messages-mget, +chat-messages-list, +messages-search, +threads-messages-list) automatically attach a reactions block and (for edited messages) update_time to each returned message — no separate im.reactions.batch_query call is needed. Pass --no-reactions to opt out. For the full contract (output shape, the im:message.reactions:read scope requirement, and the "missing field ≠ fetch failure" data rules), read references/lark-im-message-enrichment.md.

Compact message output (--concise)

Some message-listing shortcuts support --concise for compact Markdown output. Use it when the user asks for concise output or a smaller result/file; check --help for availability and do not combine it with an explicit --format, an enabled --json, or a non-empty --jq.

Opt-in resource auto-download (--download-resources)

+chat-messages-list, +messages-mget, and +threads-messages-list accept --download-resources to save eligible attachments into ./lark-im-resources/ and add a resources array to each message. It is off by default; stickers are not downloadable. A failed attachment is reported on that resource without aborting the message pull. Use +messages-resources-download for one attachment. See references/lark-im-message-enrichment.md for the output contract.

Folder resources are containers, not files — a folder file_key cannot be downloaded directly. Expand it first with lark-cli im files folder --recursive --file-key <folder_key> --srctype message --srcid <message_id>, then download the files inside with +messages-resources-download.

Card Messages (Interactive)

Before sending, replying with, or updating any interactive card (+messages-send / +messages-reply / messages.patch), you MUST read references/card/lark-im-card-create.md and follow its workflow. The card JSON passed to --msg-type interactive --content (send/reply) or messages.patch --data (update) must be the output of that workflow — never hand-write or copy a card payload.

Card messages (interactive type) are not yet supported for compact conversion in event subscriptions. The raw event data will be returned instead, with a hint printed to stderr.

interactive cards support callback events (card.action.trigger) — see references/lark-im-card-action-reply.md.

Audio Messages

--audio sends a voice message and supports only Opus audio files, for example .opus files or Ogg Opus (.ogg) files. For mp3, wav, or other non-Opus audio, either convert to .opus first and keep using --audio, or send the original file as an attachment with --file.

Sending Doc Content as a Message

When sending content fetched from a Lark doc as a message, fetch the doc with --doc-format im-markdown, then send it as a message using the --markdown format. The fetched content is already in markdown; in any content-forwarding scenario, keep the fetched original text and send it in the --markdown format. Note: if the doc contains a cite tag with type="user", keep it as-is and do not strip the tag.

Flag Types

Flags support two layers:

  • Message-layer flag: (ItemTypeDefault, FlagTypeMessage) — regular message bookmark
  • Feed-layer flag: (ItemTypeThread/ItemTypeMsgThread, FlagTypeFeed) — thread as feed-layer bookmark

Item types for feed-layer flags:

  • ItemTypeThread (4) = thread in a topic-style chat
  • ItemTypeMsgThread (11) = thread in a regular chat

Feed Shortcut

Feed shortcuts add chats to the current user's feed sidebar. They are distinct from flags:

  • Flag = bookmark on a message/thread, scoped to the user's bookmark list.
  • Feed shortcut = entry in the user's feed sidebar (currently only chats).

Key limits:

  • Only CHAT-type (feed_card_id is oc_xxx) is exposed via OpenAPI; doc/app/subscription shortcuts exist internally but are not yet whitelisted.
  • All three operations (create/remove/list) are user-identity only — they sign with user_access_token.
  • Batch size is 10 per call for create/remove; list is a one-page wrapper with opaque page_token pagination.

Shortcuts(推荐优先使用)

Shortcut 是对常用操作的高级封装(lark-cli im +<verb> [flags])。有 Shortcut 的操作优先使用。

| Shortcut | 说明 | |----------|------| | +chat-create | Create a group chat or topic chat; user/bot; --chat-mode group|topic; private/public; invites users/bots; optionally sets bot manager | | +chat-list | List chats the current user/bot is a member of; defaults to groups; pass --types=p2p,group to include p2p single chats (user-only); user/bot; supports sorting, auto-pagination, --exclude-muted (user-only) | | +chat-members-list | List members of a chat; returns separate users[] / bots[] buckets; callable as user or bot; --member-types filters which kinds to return; --page-all pagination; surfaces truncations[] when the server caps a bucket | | +chat-messages-list | List messages in a chat or P2P conversation; user/bot; accepts --chat-id or --user-id, resolves P2P chat_id, supports time range, --order asc/desc sorting, auto-pagination | | +chat-search | Search visible group chats by --query keyword and/or --member-ids; user/bot; e.g. look up chat_id by group name; supports type filters, sorting, auto-pagination, and --exclude-muted (user identity only) | | +chat-update | Update group chat name or description; user/bot; updates a chat's name or description | | +message-read-users | List users who read one message; user/bot; identity-specific scopes; supports bounded auto-pagination | | +messages-edit | Edit a message's content (text/post, including the attachment zone); bot-only (user identity is rejected by the server); PUT /open-apis/im/v1/messages/:message_id | | +messages-mget | Batch get messages by IDs; user/bot; fetches up to 50 om_ message IDs, formats sender names, expands thread replies | | +messages-read-status | Batch query whether the current user read 1–50 messages; user-only; returns readable items and invalid message IDs | | +messages-reply | Reply to a message (supports thread replies); user/bot; supports text/markdown/post/media replies, reply-in-thread, idempotency key | | +messages-resources-download | Download an image/file from a message; folders are not directly downloadable — expand with im files folder --recursive first, then download the files inside; user/bot | | +messages-search | Search messages across chats (supports keyword, sender, time range filters) with user or bot identity; filters by chat/sender/attachment/time, supports auto-pagination via --page-all / --page-limit, enriches results via batched mget and chats batch_query | | +messages-send | Send a message to a chat or direct message; user/bot; sends to chat-id or user-id with text/markdown/post/media, supports idempotency key | | +threads-messages-list | List messages in a thread; user/bot; accepts om_/omt_ input, resolves message IDs to thread_id, supports --order asc/desc sorting, auto-pagination | | +flag-create | Create a bookmark on a message; user-only; defaults to message-layer flag; use --flag-type feed for feed-layer flag (item_type auto-detected from chat mode) | | [+flag-cancel](references/la

Truncated for display — read the full file on GitHub.

Related Skills

View on GitHub
GitHub Stars17.4k
CategoryCustomer
Updated1mo ago
Forks1.4k

Languages

Go

Trust signals

100/100

From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.

No cautions