
riba2534/feishu-cli
13 skills9.2k installs15.2k starsGitHub
Install
npx skills add https://github.com/riba2534/feishu-cliSkills in this repo
1Feishu Cli ToolkitFeishu CLI Toolkit is a procedural reference for solo builders who document architecture and processes inside Feishu (Lark) using the feishu-cli. It walks through exporting whiteboard canvases as PNG images, importing PlantUML and Mermaid sources into a whiteboard with auto or explicit diagram-type mapping, and listing board nodes for inspection. It also covers embedding new whiteboard blocks into Feishu documents at a chosen parent and index. The skill fits indie teams that already keep specs in Feishu and want agent-driven or scripted diagram updates without clicking through the web UI. You invoke it when an agent needs exact command shapes, flags, and diagram-type enums—not when you need Feishu OAuth setup from scratch. Expect intermediate familiarity with CLI env vars, document IDs, and whiteboard IDs.949installs2Feishu Cli Msgfeishu-cli-msg is an agent skill for sending Feishu/Lark interactive card messages through feishu-cli’s msg send path. Solo builders automating build notifications, deploy summaries, or support escalations in Chinese enterprise chat stacks use it when the agent must deliver structured rich messages rather than plain text. The skill explains three delivery strategies—dynamic full Card JSON (recommended), Feishu-built template_id variables, and fixed card_id instances—and contrasts schema 2.0 cards (body.elements with tables, charts, forms) against deprecated v1 layouts. It intentionally does not teach card design from scratch; that belongs to feishu-cli-card in the same monorepo. Operators benefit from the header color guidance and from knowing when to pick flexible JSON versus pre-approved templates. Complexity is intermediate due to Feishu payload rules and API versioning. Pair with feishu-cli-read when messages should cite exported doc context, and with feishu-cli-card when the agent must construct new card bodies.761installs3Feishu Cli Readfeishu-cli-read is an agent skill for riba2534’s Feishu CLI that performs strictly read-only operations on Lark/Feishu cloud documents, knowledge-base pages, and spreadsheets. Solo and indie builders who keep PRDs, runbooks, or specs in Feishu can invoke it when users ask to view, read, analyze, or open content by ID, wiki node token, spreadsheet token, or URL. The skill converts remote content to Markdown stored under /tmp as an analysis intermediate—not as an automatic user-facing file drop—then uses normal agent tooling to summarize structure or answer questions. It distinguishes ordinary docx/sheet from wiki docx/sheet export paths (sheet export vs wiki export). Prerequisites include installing feishu-cli, completing auth login, and holding docx/wiki readonly scopes. For durable local copies or edits, the workflow explicitly hands off to feishu-cli-export and feishu-cli-write rather than blurring read and write concerns.761installs4Feishu Cli Writefeishu-cli-write teaches agents to create and mutate Feishu cloud documents through feishu-cli. Solo builders and tiny teams that standardize on Lark/Feishu for specs, handoffs, and customer-facing docs can spawn blank docx files, import Markdown (with image upload), and perform precise section-level edits instead of pasting into the UI. The skill documents permission grants to a resolved owner email, discourages placeholder identities, and pairs with feishu-cli-read and feishu-cli-export for a read/write/export split. It is integration-heavy: Bash invocations, optional Python UTF-8 checks, and Write/Read for staging Markdown. Best when your delivery artifact is the canonical Feishu doc your stakeholders already open daily.750installs5Feishu Cli Exportfeishu-cli-export teaches agents to pull Feishu (Lark) documents, knowledge-base nodes, and spreadsheets into local files using the feishu-cli toolchain. Solo builders and small teams use it when specs, runbooks, or PRDs live in Feishu but development happens in git: export docx and wiki nodes to Markdown, sheets to Markdown tables, and when needed kick off asynchronous PDF, Word, or Excel exports plus image and board asset downloads. The skill documents routing from URLs and tokens to the right subcommand, when to prefer export-file over Markdown, and CLI quirks such as wiki defaulting to /tmp and sheet default filenames. It is user-invocable with Bash-scoped feishu-cli permissions, making it a practical Build-phase docs bridge for Claude Code workflows centered on Chinese team collaboration stacks.735installs6Feishu Cli ImportFeishu CLI Import is an agent skill for solo builders and small teams in China or global orgs standardizing on Feishu (Lark) for specs and handbooks. Before running feishu-cli-import, the skill forces a compatibility pass on Markdown: which Mermaid diagram types auto-import versus need board import, PlantUML constraints Feishu’s renderer rejects, and how tables and images behave under doc import versus content-update with optional image upload flags. It complements feishu-cli-write for editing existing docs. The audience is developers who generate Markdown from agents or repos and need reliable diagrams and math in Feishu without manual cleanup. Intermediate complexity assumes you already use the Feishu CLI toolchain. Deliverables are a corrected Markdown file and import command choices aligned to diagram type.727installs7Feishu Cli Boardfeishu-cli-board is an agent skill that encodes Feishu whiteboard connection rules so solo builders and small teams produce legible diagrams instead of spaghetti lines. It targets anyone using Feishu CLI or agent-driven board edits who needs consistent routing when nodes fan out to children or converge from parents. Use it whenever you are generating or refactoring board JSON: the skill spells out when to draw every edge versus representative layer links, how to pick polyline versus curve versus right-angled polylines for org charts, and how to set start/end positions on normalized node edges. For builders shipping internal docs, onboarding flows, or light architecture views inside Feishu, following these rules keeps boards citable in reviews without manual cleanup. It does not replace a full design system; it is procedural layout knowledge paired with the broader Feishu CLI tooling.697installs8Feishu Cli AuthFeishu CLI Auth is an integration skill for the feishu-cli toolchain. It documents how solo builders and AI agents obtain and manage Feishu User Access Tokens using OAuth Device Flow (RFC 8628), while defaulting most commands to App Token unless the task truly needs a user identity—search, message history, approvals, meetings, or mailbox-style operations. The recommended path is scope preflight with auth check, domain login with --recommend, and status verification in JSON for automated agents. It explains Feishu error codes like 99991672 and 99991679 in context of missing scopes or expired tokens, and when to open the developer console versus re-running login. Background agents can stream JSON events so a human completes device authorization via verification_uri_complete. Use it whenever another Feishu skill fails for lack of user token or scope, not for general Lark product strategy.690installs9Feishu Cli Searchfeishu-cli-search teaches coding agents to find information inside Feishu (Lark)—cloud documents, message history, and internal apps—using the feishu-cli binary. Every search requires a User Access Token, so the skill documents a strict preflight: check scopes, log in via Device Flow when missing or expired, and only then run search commands. That matters for solo builders and small teams in China-centric orgs who want Claude Code or Cursor to answer “is there a doc about X?” without leaving the terminal. The flow covers presenting `verification_uri_complete` to the user, handling missing Open Platform permissions, and allowed bash invocations for auth and search. It is not a general web search replacement; it is Feishu-specific enterprise retrieval with explicit security boundaries around tokens and scopes.682installs10Feishu Cli PermFeishu-cli-perm is an agent skill that drives Feishu cloud document permission APIs through the feishu-cli perm commands. Solo and indie builders who run docs on Lark/Feishu use it when someone asks to share a doc, add a collaborator, grant full_access, audit who has access, or transfer ownership without manual console work. The skill embeds input checklists so agents do not mismatch doc tokens with doc types or member-type with member-id formats before running perm add. It expects App Token authentication (App ID and App Secret)—not user login—because permission management endpoints require application identity. Use it for single or bulk grants, optional Feishu notifications to invitees, and read-only versus edit versus admin scenarios. It pairs with other feishu-cli skills for content but owns the authorization surface for wikis, sheets, bitables, folders, and docx.678installs11Feishu Cli Doc Guidefeishu-cli-doc-guide is a Feishu-cli companion skill that encodes Markdown authoring rules so generated docs survive `doc import` and related Feishu/Lark APIs without broken diagrams or tables. Solo builders and small teams shipping internal runbooks, PRDs, or customer docs to Feishu install it whenever the agent drafts Markdown destined for import rather than GitHub-only rendering. It centralizes compatibility for Mermaid and PlantUML (including board import diagram types), image upload flags for import versus content-update, equation syntax, and allowed Callout variants. Invoke it before generation or when debugging render failures; pair execution with feishu-cli-import and edits with feishu-cli-write. The skill is read/grep oriented and not user-invocable by default, making it a checker layer in a documentation pipeline spanning Build docs authoring and Ship review fixes.607installs12Feishu Cli Chatfeishu-cli-chat is an agent skill that documents how to work with Feishu (Lark) external group chats through feishu-cli when platform error 232033 appears. External groups reject most in-group configuration and membership APIs unless the app has external-sharing capability enabled on the Feishu Open Platform and the app bot is actually a member of that group. The guide maps affected routes—chat info, members, announcements, tabs—and clarifies which message flows still work without those unlock steps. For indie builders targeting Chinese teams or global orgs on Feishu, this prevents agents from looping on forbidden calls and gives a concrete workflow: verify app settings, confirm bot membership with member list probes, then switch App credentials via environment variables. It is a narrow integration playbook, not general chat UX design.603installs13Feishu Cli Bitablefeishu-cli-bitable teaches agents to manage Feishu multidimensional tables (Bitable / Base) through the feishu-cli toolchain and the newer base/v3 API. Solo builders and tiny teams running sales, support, or lightweight PM on Lark can create bases, shape tables and views, upsert records, handle attachments, tune roles and advanced permissions, and maintain dashboards, forms, and workflows without clicking through every admin screen. The skill is written for Chinese-speaking operators but maps cleanly to Prism’s integration tag model: authenticate once, pass --base-token from the sheet URL, then compose bitable subcommands for CRUD and analytics. It suits agent-driven ops automation—syncing launch checklists, customer pipelines, or content calendars—while respecting Feishu’s requirement for user tokens on reads and writes. Not a replacement for Feishu’s UI for one-off edits; best when you repeat the same base operations from CI or coding agents.533installs