What does this skill do, and when should you use it?
This skill exposes the MCP tool surface of DeepChat's Feishu plugin as the primary action interface for handling Feishu/Lark documents, spreadsheets, bitables, and knowledge content. It automatically selects the appropriate tool based on user requests and extracts resource identifiers from URLs. The skill requires an active Feishu plugin preset in the current session, and tool availability depends on that preset. Its core value is allowing AI assistants to act on Feishu workspace content with natural language commands, avoiding manual app switching.
Reads, summarizes, searches, creates, updates, appends to, or organizes Feishu/Lark documents; inspects or edits Feishu/Lark spreadsheets, sheets, and bitable-like structures; handles other Feishu/Lark artifacts (like tasks, calendar, IM) when matching tools are exposed; extracts document or sheet IDs from user-provided URLs for tool calls; confirms intent on ambiguous or destructive write operations; explains the gap and suggests preset changes when no matching tool exists.
- A user who wants a quick summary of a Feishu doc without opening it.
- An analyst updating a Feishu spreadsheet cell via AI instructions.
- A project manager organizing knowledge base entries by creating or editing wiki pages.
- A writer creating new content inside a Feishu document without leaving the chat.
- An employee searching for specific information across the Feishu workspace.
How do you install this skill?
- The skill depends on Feishu plugin and MCP tools, which may involve external services; users should ensure network reachability and service availability.
- The skill does not provide rollback or undo mechanisms; destructive operations should be used with caution.
- Publisher is unverified, users should assess trust level on their own.
- Documentation lacks installation and configuration details, may be unfriendly to novices.
- Network access
- MCP Server
As part of the DeepChat repository, install the full DeepChat app and enable the Feishu plugin. The skill resides under plugins/feishu/skills/feishu-tools/ directory. Refer to DeepChat's documentation and settings UI (Settings → Skills → Install or Import) for installation and enabling. No standalone installation instructions are provided.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/ThinkInAIXYZ/deepchat.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/plugins/feishu/skills/feishu-tools" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Summarize this Feishu doc
- Update this data in the sheet
- Ensure DeepChat is installed with the Feishu plugin enabled, and that App ID, App Secret, brand, and preset are correctly configured in the plugin settings. 2. In a conversation, ask for operations on Feishu documents, spreadsheets, or knowledge bases (e.g., "Summarize this Feishu doc" or "Update this data in the sheet"). 3. The skill automatically invokes the Feishu MCP tools available in the current session. 4. Provide Feishu URLs when possible; the skill extracts required IDs. 5. If you encounter authentication or configuration errors, check the plugin settings.
What are this skill's strengths and limitations?
- Direct integration with Feishu/Lark covering documents, spreadsheets, and knowledge bases
- Real-time, concrete operations via MCP tools
- Automatic ID extraction from URLs simplifies user interaction
- Follows Agent Skills standard, portable across compatible clients
- Dependent on DeepChat's Feishu plugin and MCP server, limited standalone use
- Tool availability depends on the current preset; some operations may not be possible
- Requires pre-configuration of the Feishu plugin and appropriate permissions
- Documentation lacks detailed troubleshooting (only points to checking settings)
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Feishu Tools Skill this page | 48 · Use with care | ★ 6.4k | 3d ago | Apache-2.0 |
| LLM Wiki — a second brain for AI agents | 58 · Recommended | ★ 119 | 28d ago | MIT |
| Book-to-Skill Converter | 57 · Use with care | ★ 34k | 6d ago | MIT |
| Knowledge Base Skill | 55 · Use with care | ★ 10k | 4d ago | MIT |
| Linear Claude Skill | 55 · Use with care | ★ 134 | 2mo ago | MIT |
How did FollowSkills review this skill?
The skill file describes how to invoke Feishu MCP tools, requires confirmation for ambiguous or destructive operations, and instructs guiding users to check settings on auth errors. However, the skill itself contains no specifics about least privilege, data-flow transparency, or sensitive-data handling. No rollback mechanism or source attribution evidence is provided. Publisher is unverified but treated as unknown, not deducted for that alone. Score 14: Main risks visible, but permissions, confirmation, and isolation completeness insufficient.
The skill instructions are self-consistent, but rely on external plugins and runtime environment. No tests or failure feedback specifics are provided. On failure, only suggests user to check settings. Score 6: Happy path plausible, but edge cases, error handling, and failure feedback thin.
The skill clearly targets Feishu documents, spreadsheets, and other workspace operations, with a clear use scenario. But no capability boundaries or non-fit ranges are declared. No mention of Chinese language support or mainland-China network reachability, though Feishu is domestic, but dependency on unknown overseas services is possible. Score 10: Clear scenario, but limited evidence for boundaries and trigger conditions.
The file structure is simple with basic metadata and description, but lacks install/dependency notes, examples, FAQs, known limitations, version history, and maintenance responsibility. License and author info are visible at repo level, but not in the skill file. Score 8: Readable and usable, but with hidden assumptions and governance gaps.
The skill describes how to complete Feishu-related tasks, but no actual outputs or alternatives are provided. Value proposition is clear, but evidence for output completeness and direct usability is limited. Score 6: Completes core task, but limited evidence for output and benefit.
The skill itself contains no tests or third-party execution evidence. The repo has CI workflows, but they are for building and verifying plugin packaging, not skill behavior. Score 4: Auditable primary material exists, but coverage thin.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →