← Back to Skills Marketplace

A-Share Claw (A股龙虾)

Name: A-Share Claw (A股龙虾)
Author: youjunzhao

by ZHAO Youjun · GitHub ↗ · v0.1.1 · MIT-0

cross-platform ⚠ suspicious

111

Downloads

Stars

Active Installs

Versions

Install in OpenClaw

/install a-share-claw

Description

A-share paper-trading automation workflow for MX APIs. Use when user asks to run scheduled A-share mock trading, enforce risk limits (single position cap, to...

README (SKILL.md)

A-Share Claw

Execute this workflow for A-share paper trading with MX APIs.

Steps

Ensure environment:
- MX_APIKEY is set.
- MX_API_URL defaults to https://mkapi2.dfcfs.com/finskillshub.
Ensure runtime files under workspace:
- mx_autotrade/config.json
- mx_autotrade/run_autotrade.py
- mx_autotrade/daily_review.py
Ensure schedules (trading days):
- 09:24 run strategy
- 10:30 run strategy
- 14:30 run strategy (risk-first)
- 15:10 run daily review
Enforce risk rules:
- Single symbol position cap (maxPositionPerStock, default 15%)
- Total exposure cap (maxTotalPosition, default 60%)
- Auto-cancel stale pending orders (>20 min)
Daily output:
- Produce review JSON (mx_autotrade/reviews/review-YYYY-MM-DD.json)
- Update README day table with: Day | Balance | Daily Return (no Day0)

Quick run commands

python3 mx_autotrade/run_autotrade.py
python3 mx_autotrade/daily_review.py

Notes

Fail safely: if account is not bound, stop and report clearly.
Prefer discipline over trade frequency.

Usage Guidance

This skill is an instruction-only workflow that expects you to already have three local files (config.json, run_autotrade.py, daily_review.py) and an MX API key. Before installing or running: 1) Confirm the registry metadata omission — supply MX_APIKEY only if you trust the MX endpoint and the key is appropriate for paper trading (use a test key if available). 2) Inspect the contents of mx_autotrade/run_autotrade.py and mx_autotrade/daily_review.py line-by-line (or obtain them from a trusted source); do not run them blind because they can execute arbitrary code. 3) Verify that the risk rules (position caps, stale-order cancellation) are actually implemented in those scripts. 4) If you intend to enable scheduled or automated runs, test manually in an isolated environment or container first and ensure any scheduling mechanism is under your control. 5) Ask the skill author to correct the registry metadata to declare MX_APIKEY and to provide provenance or the actual code if they want the skill to be self-contained.

Capability Analysis

Type: OpenClaw Skill Name: a-share-claw Version: 0.1.1 The skill bundle defines a workflow for A-share paper-trading automation using the East Money (dfcfs.com) API. The instructions in SKILL.md focus on legitimate trading activities such as enforcing risk limits, scheduling strategy execution, and generating daily performance reviews. No malicious code, data exfiltration, or harmful prompt injections were found in the provided files.

Capability Assessment

⚠ Purpose & Capability

The SKILL.md clearly targets MX APIs and requires an MX_APIKEY (and optional MX_API_URL), which makes sense for an MX-integrated paper‑trading workflow — but the registry metadata lists no required environment variables or primary credential. That mismatch (metadata claims 'none' while runtime needs MX_APIKEY) is an incoherence that should be clarified before use.

⚠ Instruction Scope

The instructions tell the agent to ensure and run local scripts (mx_autotrade/run_autotrade.py and mx_autotrade/daily_review.py) and to schedule automated runs. The skill bundle does not include these files; executing them would run arbitrary Python code from the user's workspace. The SKILL.md does not include validation, safety checks, or guidance for verifying these scripts' provenance.

✓ Install Mechanism

There is no install spec (instruction-only). That minimizes installer risk because nothing from the skill itself is downloaded or written to disk.

ℹ Credentials

The runtime needs a single service credential (MX_APIKEY), which is proportionate to calling MX APIs. However, the registry metadata fails to declare this required env var/primary credential — the omission is an inconsistency and could cause silent failures or unexpected behavior if users are not alerted to provide the key.

✓ Persistence & Privilege

The skill does not request always:true and has no install actions or system-level configuration changes. Model invocation is allowed (default), which is normal; there is no evidence the skill requests elevated persistence or cross-skill config access.

How to Use

Make sure OpenClaw is installed (local or Docker)
Run the install command in chat: /install a-share-claw
After installation, invoke the skill by name or use /a-share-claw
Provide required inputs per the skill's parameter spec and get structured output

Version History

v0.1.1

Update display name with Chinese alias

v0.1.0

Initial publish

Metadata

Slug a-share-claw

Version 0.1.1

License MIT-0

All-time Installs 0

Active Installs 0

Total Versions 2

Frequently Asked Questions

What is A-Share Claw (A股龙虾)?

A-share paper-trading automation workflow for MX APIs. Use when user asks to run scheduled A-share mock trading, enforce risk limits (single position cap, to... It is an AI Agent Skill for Claude Code / OpenClaw, with 111 downloads so far.

How do I install A-Share Claw (A股龙虾)?

Run "/install a-share-claw" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is A-Share Claw (A股龙虾) free?

Yes, A-Share Claw (A股龙虾) is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does A-Share Claw (A股龙虾) support?

A-Share Claw (A股龙虾) is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created A-Share Claw (A股龙虾)?

It is built and maintained by ZHAO Youjun (@youjunzhao); the current version is v0.1.1.

More Skills