← Back to Skills Marketplace
gooderno1

LarkSync Feishu Local Cache

by gooderno1 · GitHub ↗ · v0.1.6
cross-platform ⚠ suspicious
726
Downloads
2
Stars
2
Active Installs
7
Versions
Install in OpenClaw
/install larksync-feishu-local-cache
Description
Sync Feishu docs into local cache for OpenClaw-first reading, reducing API calls and token cost. / 把飞书文档同步为本地缓存供 OpenClaw 优先读取，显著降低飞书 API 请求频率与 token 成本。
README (SKILL.md)

LarkSync Feishu Local Cache Skill

参考：

  • OPENCLAW_AGENT_GUIDE.md：面向 OpenClaw 代理的自动执行 runbook（首次授权分支、无人值守分支、失败分支）

English Overview

  • Purpose: sync Feishu documents to local markdown/files and let OpenClaw read local cache first.
  • Value: fewer Feishu API calls, lower token consumption, and more stable retrieval latency.
  • Default mode: download_only with low-frequency schedule (daily by default).
  • Advanced mode: bidirectional and upload_only are supported when explicitly requested.

Typical intents (EN)

  • "Set up a daily 01:00 sync from this Feishu folder to local cache."
  • "Check my current LarkSync auth and task status first."
  • "Switch this task to bidirectional sync and explain the risk."

价值主张

  • 让 OpenClaw 的高频文档问答“走本地、不走云端接口”。
  • 把 API 调用从“每次问答触发”变成“低频定时同步触发”。
  • 在不改用户飞书使用习惯的前提下，直接获得更稳、更省的文档检索链路。

适用目标

  • 目标：减少 OpenClaw 直接调用飞书 API 的频率，优先读取本地 Markdown/附件。
  • 默认策略：download_only + 每日低频同步（可自定义时间与周期）。
  • 进阶策略：支持 bidirectional（双向）和 upload_only，但仅在用户明确要求时启用。
  • Windows 侧建议直接使用 LarkSync 安装包版运行服务，不需要源码构建。

触发意图（示例）

  • “把飞书生产测试文件夹每天 01:00 同步到本地，给我配置好。”
  • “先检查 LarkSync 当前授权和同步任务状态。”
  • “我想把这个目录切成双向同步，先评估风险再改。”

默认执行流程

  1. 检查本地 LarkSync 服务与授权状态。
  2. 配置低频下载策略（默认每天一次）。
  3. 创建同步任务（默认 download_only）。
  4. 按需触发一次立即同步，建立本地缓存基线。
  5. 后续回答用户文档问题时，优先读取本地同步目录。

命令入口

使用以下脚本作为统一入口（返回 JSON，便于自动化编排）：

python integrations/openclaw/skills/larksync_feishu_local_cache/scripts/larksync_skill_helper.py check
python integrations/openclaw/skills/larksync_feishu_local_cache/scripts/larksync_skill_helper.py configure-download --download-value 1 --download-unit days --download-time 01:00
python integrations/openclaw/skills/larksync_feishu_local_cache/scripts/larksync_skill_helper.py create-task --name "OpenClaw 每日同步" --local-path "D:\\Knowledge\\FeishuMirror" --cloud-folder-token "\x3CTOKEN>" --sync-mode download_only
python integrations/openclaw/skills/larksync_feishu_local_cache/scripts/larksync_skill_helper.py run-task --task-id "\x3CTASK_ID>"
python integrations/openclaw/skills/larksync_feishu_local_cache/scripts/larksync_skill_helper.py bootstrap-daily --local-path "D:\\Knowledge\\FeishuMirror" --cloud-folder-token "\x3CTOKEN>" --sync-mode download_only --download-value 1 --download-unit days --download-time 01:00 --run-now

WSL 场景（OpenClaw 在 WSL，LarkSync 在 Windows）推荐入口：

# 先做地址诊断（逐项显示 localhost / host.docker.internal / gateway / resolv nameserver）
python integrations/openclaw/skills/larksync_feishu_local_cache/scripts/larksync_wsl_helper.py diagnose

# 直接执行原有命令（自动探测可达地址；远程地址自动补 --allow-remote-base-url）
python integrations/openclaw/skills/larksync_feishu_local_cache/scripts/larksync_wsl_helper.py check
python integrations/openclaw/skills/larksync_feishu_local_cache/scripts/larksync_wsl_helper.py bootstrap-daily --local-path "/mnt/d/Knowledge/FeishuMirror" --cloud-folder-token "\x3CTOKEN>" --sync-mode download_only --download-value 1 --download-unit days --download-time 01:00 --run-now

若全部候选地址不可达，优先确认 Windows 侧：

  • LarkSync 已启动；
  • 若手动设置过 LARKSYNC_BACKEND_BIND_HOST=127.0.0.1，请改回 0.0.0.0 或移除该变量；
  • 防火墙已放行 WSL 网段到 TCP 8000。

安全边界：

  • 未探测到 Windows 侧可达服务时，脚本会输出诊断信息并停止，不会自动在 WSL 安装依赖或启动后端。
  • 请先在 Windows 侧启动 LarkSync（安装包版或开发模式），确认 :8000 可达后再重试。
  • 注意：飞书 OAuth 首次授权仍需用户完成；授权完成后可进入日常低频同步运行。

默认安全策略：

  • base-url 仅允许 localhost/127.0.0.1/::1，避免把云目录 token 发送到未知远端地址。
  • 只有在明确可信内网场景，才允许开启远程地址：
python integrations/openclaw/skills/larksync_feishu_local_cache/scripts/larksync_skill_helper.py --base-url "https://larksync.internal.example" --allow-remote-base-url check

约束与安全边界

  • 未通过 check 之前，不执行任务创建或策略变更。
  • 未经用户明确同意，不把 download_only 自动切到 bidirectional。
  • 若用户要开启双向，必须先告知风险：
    • 本地误改可能上云；
    • 首次建任务可能触发下行/上行扫描；
    • 建议先在测试目录验证。

失败处理

  • 若接口返回 401/403：提示重新授权飞书并检查“用户身份权限”。
  • 若创建任务冲突（409）：自动复用同路径+同云目录的现有任务并回显任务 ID。
  • 若后端不可达：优先提示启动 LarkSync 安装包版托盘程序（开发者再使用 npm run dev）。

输出规范

  • 对用户：简明中文结论 + 下一步操作。
  • 对系统：保留 helper 脚本 JSON 原始输出，便于追踪与审计。
Usage Guidance
This skill appears to do what it claims: configure and drive a local LarkSync backend to mirror Feishu docs. Before installing or allowing an agent to run it: 1) Ensure you actually have a local LarkSync backend running and reachable (http://localhost:8000) — run the provided check first. 2) Treat the cloud_folder_token as a secret: only provide it when base_url is local/trusted. The script will refuse non-loopback base_url unless you explicitly permit it (--allow-remote-base-url). 3) If running under WSL, review the diagnose output (the helper probes /proc/net/route and /etc/resolv.conf) to confirm the chosen base_url; these probes are normal for endpoint discovery but reveal network topology. 4) Prefer the default download_only mode until you understand bidirectional risks (local edits could be uploaded). 5) If you plan to allow autonomous agent invocation, be aware the agent can call these helpers and therefore trigger sync actions — only enable that for trusted agents and environments.
Capability Analysis
Type: OpenClaw Skill Name: larksync-feishu-local-cache Version: 0.1.6 The skill is designed for a legitimate purpose (syncing Feishu docs to local cache) and includes several explicit security controls, such as defaulting to loopback-only connections and warning about token leakage risks for remote URLs. However, the `larksync_skill_helper.py` script allows `--allow-remote-base-url` to bypass the loopback restriction, and the `larksync_wsl_helper.py` script automatically injects this flag for non-loopback addresses in WSL scenarios. While documented, this capability (in `scripts/larksync_skill_helper.py` and `scripts/larksync_wsl_helper.py`) allows the `cloud_folder_token` to be sent to arbitrary remote endpoints if the agent is compromised or misconfigured, representing a significant vulnerability for credential exfiltration. The `local_path` argument also expands user paths, which could lead to path traversal if not properly handled by the backend or if the agent is prompted maliciously. These are risky capabilities that, if exploited, could lead to harm, classifying the skill as suspicious rather than benign, despite the explicit warnings and other safeguards.
Capability Assessment
✓ Purpose & Capability
Name/description (sync Feishu docs into a local cache) match the delivered artifacts: helper scripts that call a LarkSync HTTP API and docs describing download_only/bidirectional modes. No unrelated credentials or cloud SDKs are requested. Code focuses on creating/configuring/scheduling tasks on a LarkSync backend, which is coherent with the stated purpose.
✓ Instruction Scope
SKILL.md instructs running the provided helper scripts and explicitly shows where to supply a cloud_folder_token. The helpers only interact with a LarkSync API (health, auth/status, config, /sync/tasks) and perform WSL network probes to discover reachable base URLs. There are explicit safety checks (default base_url limited to loopback unless --allow-remote-base-url is provided) and a required `check` step before task creation. The instructions do not ask the agent to read unrelated user files or exfiltrate data.
✓ Install Mechanism
This is effectively instruction-only (no install spec). The package includes two helper Python scripts but contains no download/install steps, no archive extraction, and no network-based installer. No external binary installs are scripted by the skill itself.
ℹ Credentials
The skill requests no environment variables or platform credentials in metadata. It expects caller-provided tokens via CLI (cloud_folder_token) which is appropriate for the task. Note: cloud_folder_token is sensitive and will be sent to the configured base_url; the code restricts base_url to localhost by default and requires an explicit flag to use remote endpoints.
✓ Persistence & Privilege
The skill does not request always:true, does not modify other skills, and has no persistent install behavior. Helpers do not auto-install dependencies or launch services; WSL wrapper explicitly documents it won't start the backend automatically.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install larksync-feishu-local-cache
  3. After installation, invoke the skill by name or use /larksync-feishu-local-cache
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.6
fix(security): remove WSL auto-install and auto-start behaviors
v0.1.5
fix(wsl-runtime): sanitize pythonpath for autonomous local backend startup
v0.1.4
feat(wsl-agent): autonomous WSL fallback runtime + agent runbook
v0.1.3
feat(wsl): add WSL wrapper helper with multi-endpoint diagnostics and auto remote allow flag
v0.1.2
docs: add bilingual (EN/CN) introduction for ClawHub listing
v0.1.1
安全修复：默认仅允许 localhost base-url，远程地址需显式开启
v0.1.0
首次发布：飞书低频同步到本地缓存
Metadata
Slug larksync-feishu-local-cache
Version 0.1.6
License —
All-time Installs 2
Active Installs 2
Total Versions 7
Frequently Asked Questions

What is LarkSync Feishu Local Cache?

Sync Feishu docs into local cache for OpenClaw-first reading, reducing API calls and token cost. / 把飞书文档同步为本地缓存供 OpenClaw 优先读取，显著降低飞书 API 请求频率与 token 成本。 It is an AI Agent Skill for Claude Code / OpenClaw, with 726 downloads so far.

How do I install LarkSync Feishu Local Cache?

Run "/install larksync-feishu-local-cache" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is LarkSync Feishu Local Cache free?

Yes, LarkSync Feishu Local Cache is completely free (open-source). You can download, install and use it at no cost.

Which platforms does LarkSync Feishu Local Cache support?

LarkSync Feishu Local Cache is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created LarkSync Feishu Local Cache?

It is built and maintained by gooderno1 (@gooderno1); the current version is v0.1.6.

More Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463283 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259410 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200423 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191115 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190156 · by oswalpalash

💬 Comments