Lark Skills Roundup: Slides XML, Task GUIDs, Whiteboard CLI, OKR Locks, Minutes Shim
Five Feishu lark skills, read from their files: a 15-line shim, task GUID traps, slides XML gates, whiteboard npm deps, OKR score locks — 718–738K installs.
Manage Lark/Feishu Base records, tables, and fields via AI. Query data, update rows, and automate workflows without leaving your coding assistant.
Search, read, and manage Lark/Feishu wiki pages and knowledge bases via AI. Find documents instantly, organize knowledge, and connect wiki content to your code.
Feishu’s lark-approval skill handles the full approval lifecycle from the CLI — file instances, approve, transfer, rollback, remind. 728.9K installs, rank 127.
Feishu’s lark-apps skill builds and hosts Miaoda apps from the CLI — local git or cloud AI sessions, releases, env vars, databases. 632,894 installs, #45.
Feishu’s lark-attendance skill is the thinnest file in the lark pack: one endpoint, permanently fixed params, self-only by construction. 728.7K installs.
Feishu’s lark-contact resolves names, emails, and bots into open_ids (449.8K installs) — with two strictly separate user and bot command paths.
Feishu’s lark-im sends messages, manages group chats, and pins feed shortcuts from the CLI — 736.5K installs plus a hard user-vs-bot identity split per call.
Feishu’s lark-mail drafts, sends, and triages email via lark-cli: draft-by-default writes, confirm-before-send gates, prompt-injection guards — 737.8K installs.
Lark’s lark-markdown skill edits native .md files in Drive — fetch, create, overwrite, patch, diff — and never converts them to docx. 684,911 installs.
Lark’s lark-vc-agent ranks #38 with 559K installs, but its SKILL.md is a 466-byte shim routing every request to lark-meeting.
Read, create, and edit Lark/Feishu docs using AI. Generate documents, batch-update content, and integrate document workflows into your development pipeline.
Five Feishu lark skills, read from their files: a 15-line shim, task GUID traps, slides XML gates, whiteboard npm deps, OKR score locks — 718–738K installs.
Feishu ships two data skills that look alike: lark-sheets (grid cells, formulas) and lark-base (typed records, views, forms). This maps when to use each.
lark-event (450.1K installs) streams Feishu IM, VC, and Minutes events as NDJSON. The subprocess contract — ready markers, stdin EOF, exit reasons — matters.
Procedural walkthrough of Feishu’s lark-calendar skill: classify new vs edit, lock the event_id, branch on time clarity, book rooms as attendees.
Straight answers on Feishu’s lark-meeting skill: meeting_id vs minute_token vs note_id, user vs bot identity, transcripts from Minutes — with 126.9K installs.
Set up Lark Knowledge AI in six steps — structure Wiki knowledge spaces, enable AI search and summaries, and connect the lark-wiki skill to your coding agent.
I handed my AI agent control over our team’s approval workflow for five days. Expense reports, leave requests, purchase orders — all processed through the lark-approval AI skill. Here is what went right, what broke, and whether I would do it again.
I handed my AI agent the keys to our team’s Lark documentation for a full work week — reading, writing, organizing, and searching across 40+ docs. Here is what worked, what broke, and how the LarkDoc skill changed how I think about AI-assisted documentation.
Compare four AI workflow automation skills — Microsoft Foundry, Azure AI, Lark Doc, and Lark Base — to find the right fit for automating your team’s processes.
Compare Lark Doc and Lark Wiki — two knowledge management skills in the larksuite CLI. Honest verdict on which one to install based on your workflow.
Step-by-step guide to installing the LarkDoc skill, authenticating with Lark/Feishu, and using AI agents to read, create, and edit documents through natural language.