← Back to Skills Marketplace
tianyedavid

akshare-local-workbench

by DavidTian · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
48
Downloads
1
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install akshare-local-workbench
Description
Install, run, maintain, and troubleshoot a bundled local single-user AKShare financial data workbench. Use this skill when OpenClaw needs to create the workb...
README (SKILL.md)

AKShare Local Workbench

Use this skill to create and operate a local AKShare data workbench from bundled source files. This standalone ClawHub package includes the app source under assets/akshare-workbench, so init-project does not clone GitHub.

Treat the app as a local single-user tool. Prefer caching, slow request pacing, and targeted indicators over high-frequency or bulk data-source access.

Quick Start

Create the workbench project from bundled assets:

python3 /path/to/akshare-local-workbench/scripts/workbench_ctl.py init-project --target akshare-workbench

Then install dependencies and start:

cd akshare-workbench
python3 /path/to/akshare-local-workbench/scripts/workbench_ctl.py setup
python3 /path/to/akshare-local-workbench/scripts/workbench_ctl.py start

When installed through OpenClaw/ClawHub, use the skill's installed script path. If the app already exists, run commands from inside the app project or pass:

--root /path/to/akshare-workbench

Main Commands

python3 /path/to/akshare-local-workbench/scripts/workbench_ctl.py init-project --target akshare-workbench
python3 /path/to/akshare-local-workbench/scripts/workbench_ctl.py doctor
python3 /path/to/akshare-local-workbench/scripts/workbench_ctl.py setup
python3 /path/to/akshare-local-workbench/scripts/workbench_ctl.py start
python3 /path/to/akshare-local-workbench/scripts/workbench_ctl.py status
python3 /path/to/akshare-local-workbench/scripts/workbench_ctl.py restart
python3 /path/to/akshare-local-workbench/scripts/workbench_ctl.py test
python3 /path/to/akshare-local-workbench/scripts/workbench_ctl.py clear-cache
python3 /path/to/akshare-local-workbench/scripts/workbench_ctl.py stop

The frontend defaults to http://127.0.0.1:5173; backend health is http://127.0.0.1:8000/api/health.

Request-Limit Guidance

The controller starts the backend with conservative defaults:

AKSHARE_MAX_CONCURRENT=1
AKSHARE_EASTMONEY_INTERVAL_SECONDS=12
AKSHARE_EASTMONEY_HTTP_RETRIES=2
AKSHARE_EASTMONEY_CALL_ATTEMPTS=1
AKSHARE_EASTMONEY_CACHE_TTL_SECONDS=1800
AKSHARE_RESULT_CACHE_TTL_SECONDS=900
AKSHARE_ENRICH_NAMES=0

Ordinary "提取" uses local cache. "强制刷新" bypasses cache and should be used sparingly.

For Eastmoney/东方财富 RemoteDisconnected or Connection aborted errors:

  1. Wait several minutes.
  2. Avoid repeated forced refreshes.
  3. Prefer single-symbol indicators over full-market real-time endpoints.
  4. Try AKSHARE_PROXY_MODE=direct or AKSHARE_PROXY_MODE=system.
  5. Raise AKSHARE_EASTMONEY_INTERVAL_SECONDS to 20 or 30.

Do not use IP rotation, high-concurrency scraping, or attempts to bypass upstream restrictions. For production, commercial, high-frequency, or bulk use, use an official or licensed data feed.

Maintenance

Read references/maintenance.md for bundled asset details, common commands, catalog maintenance, troubleshooting, and open-source/disclaimer notes.

Usage Guidance
Install only if you are comfortable running a local financial-data workbench that installs Python/npm dependencies and contacts public data providers. If you enable AI, use a provider you trust, avoid entering sensitive financial/account information into chat, and treat the local ai_config.json file as a secret because it stores the API key in plaintext.
Capability Tags
cryptorequires-sensitive-credentials
Capability Assessment
Purpose & Capability
The bundled files match the stated purpose: a local single-user AKShare workbench for selecting indicators, fetching public financial data, previewing/exporting results, and optionally using an AI assistant to map natural-language requests to known indicators.
Instruction Scope
Runtime instructions are user-directed and scoped to initializing, setting up, starting, stopping, testing, and maintaining the local workbench; the skill also advises conservative request pacing and avoiding bypass of upstream data-provider limits.
Install Mechanism
The controller copies bundled assets, creates a Python virtual environment, installs declared Python and npm dependencies, and starts local backend/frontend processes bound to 127.0.0.1; destructive replacement only occurs when the user explicitly passes --force.
Credentials
Network access to AKShare-backed public data sources and a user-configured OpenAI-compatible model endpoint is proportionate to the workbench purpose, but users should understand that AI prompts and recent chat history go to the configured provider.
Persistence & Privilege
The app persists result cache files locally with TTLs and stores AI model configuration, including the API key, in a local JSON file; this is disclosed in project notes and not returned by the config API, but it is plaintext and not OS-keychain protected.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install akshare-local-workbench
  3. After installation, invoke the skill by name or use /akshare-local-workbench
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of akshare-local-workbench. - Provides local install, operation, and maintenance of a single-user AKShare financial data workbench from bundled assets—no GitHub cloning required. - Includes command-line controller for project setup, dependency install, starting/stopping services, testing, cache management, and troubleshooting. - Supports editable AKShare YAML indicator catalog and manages conservative defaults to reduce Eastmoney/东方财富 rate-limit errors. - Exposes environment variables to customize ports, proxy mode, root path, and request pacing. - Emphasizes local caching, slow pacing, and responsible use for single-user scenarios.
Metadata
Slug akshare-local-workbench
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is akshare-local-workbench?

Install, run, maintain, and troubleshoot a bundled local single-user AKShare financial data workbench. Use this skill when OpenClaw needs to create the workb... It is an AI Agent Skill for Claude Code / OpenClaw, with 48 downloads so far.

How do I install akshare-local-workbench?

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

Is akshare-local-workbench free?

Yes, akshare-local-workbench is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does akshare-local-workbench support?

akshare-local-workbench is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created akshare-local-workbench?

It is built and maintained by DavidTian (@tianyedavid); the current version is v1.0.0.

💬 Comments