← Back to Skills Marketplace
QuantClaw
by
natanhoffmann
· GitHub ↗
· v1.0.6
· MIT-0
382
Downloads
0
Stars
0
Active Installs
7
Versions
Install in OpenClaw
/install quant-claw
Description
LLM-driven crypto perpetual futures trading on Bybit. Analyzes 200+ market features across 8 timeframes with strategy-aware output, then you decide whether t...
Usage Guidance
This skill appears to be what it claims (a Bybit trading CLI wrapper) but take these precautions before installing or supplying API keys:
- Prefer Bybit demo/paper-trading API keys first. Never provide live keys with withdrawal permissions.
- The quantclaw CLI is not installed automatically; you must purchase/download it from Gumroad. Verify the provided SHA256 checksum and review the source code before running.
- The SKILL.md relies on an interactive confirmation to prevent autonomous LIVE trading — that is an instruction, not a platform-enforced control. If you enable autonomous agent invocation and give live keys, the agent could execute trades unless you strictly control when and how the skill is invoked.
- If you are not comfortable verifying third-party Python code or managing keys, avoid installing or only use the demo environment behind a separate, limited API key.
If you want a stronger assessment, provide the downloaded quantclaw source (or a link to an official release repo) so the code and any network/call patterns can be reviewed.
Capability Analysis
Type: OpenClaw Skill
Name: quant-claw
Version: 1.0.6
The bundle is a documentation-only stub for a commercial crypto trading CLI tool (QuantClaw). It defines an interface for an AI agent to perform market analysis and execute trades on Bybit, but does not include the actual executable code, which requires a manual external purchase from Gumroad. The documentation (SKILL.md) includes explicit safety instructions for the agent, such as defaulting to demo mode and requiring interactive confirmation for live trading, and shows no signs of malicious intent, obfuscation, or data exfiltration.
Capability Assessment
Purpose & Capability
Name/description (Bybit perpetual futures trading) align with requested env vars (BYBIT_API_KEY, BYBIT_API_SECRET) and required binaries (uv to run a Python environment, quantclaw CLI). Requesting Bybit keys is expected; requiring the quantclaw CLI is consistent with a paid product distribution.
Instruction Scope
SKILL.md instructs the agent to run the quantclaw CLI for analysis and trading and emphasizes DEMO-by-default and an interactive confirmation for LIVE. The instructions do not request unrelated files or extra credentials, but they rely on the agent and the CLI to honor the 'interactive confirmation' restriction — this is an honor-system safety control, not an enforceable platform-level guarantee.
Install Mechanism
Installation for uv via Homebrew or pip is low-risk. The quantclaw CLI distribution is 'manual' via a Gumroad purchase; manual download of third-party code is higher risk than package-managed installs. The SKILL.md states the purchase includes source and a SHA256 checksum, which mitigates risk if you actually verify the files before running.
Credentials
Only Bybit API key and secret are required, which is proportionate for a trading skill. The doc correctly recommends demo keys and warns about not using keys with withdrawal permissions. That recommendation is appropriate and necessary.
Persistence & Privilege
always=false (not force-included), which is appropriate. However, the platform default allows autonomous invocation; if you install the skill and provide live trading keys the agent could execute the quantclaw CLI and place orders unless the agent strictly follows the SKILL.md's interactive-confirmation rule. That enforcement is at the skill/agent level, not guaranteed by the registry.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install quant-claw - After installation, invoke the skill by name or use
/quant-claw - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.6
Clarify LIVE mode requires interactive user confirmation, non-bypassable by agent or env flags
v1.0.5
Fix domain mismatch, replace curl|sh with brew/pip, add source code + SHA256 provenance note
v1.0.4
Update Gumroad URL to anomalysystems.gumroad.com
v1.0.3
Fix: DEMO mode is paper trading, not testnet
v1.0.2
Add homepage, explain uv dependency — resolves security scan flags
v1.0.1
Fix metadata: add install spec, env docs, safety note
v1.0.0
QuantClaw 1.0.0
- Initial release of QuantClaw.
https://quantclaw.gumroad.com/l/wugjom
• LLM-driven crypto perpetual futures trading on Bybit.
• 200+ indicators across 8 timeframes — RSI, MACD, ADX, CVD, funding rate, OI, volume profile, market structure, and more.
• 4 strategies: SWING, FAST, MEDIUM, POSITION — each tuned to a different hold duration.
• Ask in plain English: "Is BTC ready to long?" — no code, no config, just signal.
• Risk gates built in — automatically flags untradeable setups before you size in.
• Runs on Bybit demo or live accounts.
- Proprietary license; full version available for purchase at the provided Gumroad link.
- One-time purchase, no subscriptions required.
Metadata
Frequently Asked Questions
What is QuantClaw?
LLM-driven crypto perpetual futures trading on Bybit. Analyzes 200+ market features across 8 timeframes with strategy-aware output, then you decide whether t... It is an AI Agent Skill for Claude Code / OpenClaw, with 382 downloads so far.
How do I install QuantClaw?
Run "/install quant-claw" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is QuantClaw free?
Yes, QuantClaw is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does QuantClaw support?
QuantClaw is cross-platform and runs anywhere OpenClaw / Claude Code is available (darwin, linux).
Who created QuantClaw?
It is built and maintained by natanhoffmann (@natanhoffmann); the current version is v1.0.6.
More Skills