← Back to Skills Marketplace
murphys7017

和风天气查询功能

by murphys7017 · GitHub ↗ · v1.0.0
cross-platform ⚠ suspicious
2423
Downloads
2
Stars
16
Active Installs
1
Versions
Install in OpenClaw
/install qweather
Description
天气查询:使用和风天气(JWT+Host)获取实时天气与未来预报;支持城市名/LocationID/经纬度;缺省地点可用 QWEATHER_DEFAULT_LOCATION。
Usage Guidance
This skill mostly does what it says (weather lookups), but it expects enterprise-style JWT credentials and a private key while the registry metadata declares no required secrets. Before installing, ask the author to: 1) explicitly list required env vars and the primary credential in the registry (PROJECT_ID, CREDENTIALS_ID, QWEATHER_PRIVATE_KEY_PATH or QWEATHER_API_HOST); 2) remove any checked-in private key files (or confirm the file is only a harmless placeholder) and instead require the private key path be provided via a secure env var or secret store; 3) confirm the API_HOST and other config are not hard-coded placeholders; and 4) document why QWEATHER_FREE_API_KEY and QWEATHER_DEFAULT_LOCATION are needed. If you cannot verify those, avoid installing or running the skill with real credentials on sensitive systems.
Capability Analysis
Type: OpenClaw Skill Name: qweather Version: 1.0.0 The skill is designed to query weather data using the QWeather API, which requires JWT authentication. This involves reading a private key from a file (`lib/ed25519-private.txt` is a template, but the code in `lib/qweather_jwt_session.js` and `lib/gen_qweather_token.js` uses `fs.readFileSync` to load it). While this file access is necessary for the skill's stated purpose of JWT signing, it represents a high-risk capability (direct access to a private key file) that could be exploited if the `PRIVATE_KEY_PATH` configuration were maliciously altered. Additionally, the `gen_qweather_token.js` utility script logs the generated JWT token to the console, which could be a minor information leak in certain environments.
Capability Assessment
Purpose & Capability
The skill's code and SKILL.md implement weather lookups via QWeather (and fallbacks to Open-Meteo), which matches the declared purpose. However, the implementation expects a QWeather JWT flow (PROJECT_ID, CREDENTIALS_ID, PRIVATE_KEY_PATH, API_HOST) but the registry metadata lists no required env vars/credentials. That mismatch (hidden/undeclared credentials) is unexpected for a simple weather skill.
Instruction Scope
SKILL.md instructs the agent to use the bundled tools (location lookup, now, forecast) and references env vars such as QWEATHER_DEFAULT_LOCATION and QWEATHER_PRIVATE_KEY_PATH. The code reads a private key file (CONFIG.PRIVATE_KEY_PATH) from disk, signs JWTs, and calls external APIs. The instructions are not overly broad, but they rely on reading local key material and several environment variables that are not declared in the skill metadata — this expands the runtime surface without making that explicit.
Install Mechanism
There is no install spec and no downloads. The skill is instruction+code only and uses standard Node libraries (crypto, axios). No external installers or remote archives are used.
Credentials
The code requires private key material and QWeather credentials (PROJECT_ID, CREDENTIALS_ID, PRIVATE_KEY_PATH, API_HOST) to use the enterprise JWT-backed API, and also checks process.env.QWEATHER_FREE_API_KEY and QWEATHER_DEFAULT_LOCATION in fallback code. None of these are declared in the skill's required env/credential metadata. Additionally the repo includes lib/ed25519-private.txt which does not contain a key but an openssl command wrapped in PEM markers — its presence is unusual and could be a placeholder or a mistaken checked-in secret. Requiring private keys or API keys without declaring them is disproportionate and surprising to users.
Persistence & Privilege
always is false and there is no code that attempts to modify other skills or global agent settings. The skill does read files from its own package (the configured PRIVATE_KEY_PATH) but does not request persistent system privileges.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install qweather
  3. After installation, invoke the skill by name or use /qweather
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of weather-query skill for QWeather integration. - Provides real-time weather and forecasts using QWeather API (JWT+Host). - Supports queries by city name, LocationID, or latitude/longitude. - Allows “today/tomorrow/the day after/future N days” forecast expressions. - Automatically resolves location input or uses QWEATHER_DEFAULT_LOCATION if set. - Returns comprehensive weather details: condition, temperature, humidity, wind, observation time, and daily forecast highlights. - Outputs clear errors with env variable guidance on API failure.
Metadata
Slug qweather
Version 1.0.0
License
All-time Installs 18
Active Installs 16
Total Versions 1
Frequently Asked Questions

What is 和风天气查询功能?

天气查询:使用和风天气(JWT+Host)获取实时天气与未来预报;支持城市名/LocationID/经纬度;缺省地点可用 QWEATHER_DEFAULT_LOCATION。 It is an AI Agent Skill for Claude Code / OpenClaw, with 2423 downloads so far.

How do I install 和风天气查询功能?

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

Is 和风天气查询功能 free?

Yes, 和风天气查询功能 is completely free (open-source). You can download, install and use it at no cost.

Which platforms does 和风天气查询功能 support?

和风天气查询功能 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created 和风天气查询功能?

It is built and maintained by murphys7017 (@murphys7017); the current version is v1.0.0.

💬 Comments