crypto-watch-skill
PassAudited by ClawScan on May 1, 2026.
Overview
This appears to be a coherent crypto market-data and technical-analysis skill that uses public APIs and does not request credentials or perform trades, but users should notice its external calls, Python dependencies, and actionable trading-plan guidance.
Before installing, understand that this skill fetches public market data from external providers and can produce detailed trading-plan style outputs. It does not show evidence of asking for keys, placing trades, or persisting data, but you should use an isolated Python environment and treat any market recommendations as non-authoritative.
Findings (4)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The agent may preferentially use this skill whenever a crypto or market-data question comes up.
The skill gives strong trigger instructions that can steer an agent to use it for crypto-related queries; this is disclosed and aligned with the skill purpose.
MUST load this Skill when: - Querying any crypto price ... - Technical analysis requests ... - Market sentiment analysis
Install it only if you want crypto-related queries to be routed through this market-data workflow.
Crypto symbols and analysis parameters may be sent to OKX or other disclosed public data providers.
The CLI sends user-selected asset symbols, timeframes, and limits to external market-data APIs, which is expected for this skill but still means queries leave the local environment.
url = "https://www.okx.com/api/v5/market/candles" ... params = {"instId": inst_id, "bar": bar, "limit": limit} ... resp = requests.get(url, params=params, timeout=DEFAULT_TIMEOUT)Use it for market-data queries you are comfortable sending to public API providers; do not include private keys or account information.
Installing dependencies could pull newer package versions with changed behavior or newly introduced vulnerabilities.
Dependencies are specified with lower-bound ranges, so a future install may resolve to newer package versions than those originally tested.
pandas>=2.0.0 numpy>=1.24.0 requests>=2.31.0 urllib3>=2.0.0
Install in an isolated environment and consider pinning dependency versions if you need reproducible behavior.
A user might treat generated trade plans as authoritative financial advice.
The strategy reference encourages actionable trading-plan language and leverage suggestions, which can carry financial impact even though the skill says it does not execute trades.
Every analysis MUST include: 1. Direction: Long / Short / Neutral ... 5. Position Size ... Recommended Leverage ... 8x ~ 10x
Treat outputs as educational market analysis, verify data independently, and avoid relying on the agent alone for leveraged trading decisions.