perplexity
Use Perplexity API for web-grounded AI search. Use when user needs up-to-date information with source citations, factual queries about current events, or research-style answers. Default when user mentions Perplexity or needs current information with references.
Why use this skill?
Integrate Perplexity AI into OpenClaw for secure, web-grounded research with source citations. Get accurate, real-time answers today.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/haru3613/perplexity-safeWhat This Skill Does
The Perplexity skill enables the OpenClaw agent to perform sophisticated, web-grounded searches using the Perplexity API. Unlike standard web scrapers, this tool leverages advanced large language models to synthesize real-time information from across the internet, providing highly accurate answers complete with verifiable source citations. The skill is architected as a security-hardened replacement for legacy search implementations, ensuring that user inputs are sanitized and safely transmitted to the backend API without risk of command injection or malicious execution. By utilizing specialized models like sonar-pro or sonar-reasoning, the agent can handle everything from simple factual verification to complex multi-step research tasks, ensuring that the information provided is both current and reliable.
Installation
To integrate this skill, navigate to your OpenClaw environment and use the command: clawhub install openclaw/skills/skills/haru3613/perplexity-safe. After installation, ensure you have a valid Perplexity API key. You can authenticate either by creating a config.json file inside the skill directory containing your apiKey or by exporting the PERPLEXITY_API_KEY environment variable directly in your shell configuration. The system prioritizes the local config file for added security and portability.
Use Cases
Use this skill whenever accuracy and recency are paramount. It is the ideal tool for tracking breaking news, monitoring stock market fluctuations, or performing deep-dive research into academic or technical topics where static model knowledge might be outdated. It serves as an excellent companion for developers needing up-to-date library documentation or researchers requiring a list of primary sources to validate claims.
Example Prompts
- "What are the latest performance benchmarks for the newest NVIDIA GPU released this month? Please include citations."
- "Compare the pros and cons of using Rust versus Go for microservices, citing current industry expert articles."
- "Search for the current status of the international climate summit and summarize the key agreements reached so far."
Tips & Limitations
To optimize results, match the model to the complexity of the query. For fast, factual answers, the default sonar model is sufficient. If you require deep reasoning or comprehensive report-style output, switch to sonar-reasoning or sonar-reasoning-pro. Always specify the -f markdown flag if you intend to display the output within an OpenClaw chat interface to ensure proper formatting of the citations. Note that while this tool provides web-grounded data, it is subject to the rate limits and costs associated with your specific Perplexity API plan. Avoid using this for simple local file navigation; use built-in system tools for those tasks to conserve API tokens.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-haru3613-perplexity-safe": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: network-access, external-api