presearch-search

What This Skill Does

The Presearch Search skill provides OpenClaw agents with access to a decentralized, privacy-focused search engine. Unlike traditional search providers that monetize user data and rely on centralized algorithms, Presearch utilizes a distributed network of nodes to deliver uncensored, objective search results. This skill allows your AI agent to fetch real-time data from the web, verify facts, explore trending topics, and conduct research without compromising user privacy. The integration supports standard query parameters like language filtering, time-based results, and pagination, making it a robust tool for data gathering and information retrieval tasks.

Installation

To integrate this skill into your environment, use the OpenClaw command-line interface. Ensure your local configuration is authenticated, then run the following command in your terminal:

clawhub install openclaw/skills/skills/nosytlabs/presearch

Once installed, initialize the skill within your agent by providing your Bearer API key. The skill will automatically handle connection pooling and rate limiting to ensure optimal performance within the 100 requests/minute threshold.

Use Cases

This skill is ideal for agents that require up-to-date information without the bias of major tech platforms. Use it for:

• Fact-checking claims or verifying current events in real-time.
• Gathering diverse perspectives on controversial or sensitive topics where mainstream search results might be filtered.
• Conducting deep-dive research for technical documentation or academic writing.
• Aggregating external resources for automated reports or summaries.

Example Prompts

1. "Use Presearch to find the latest updates on the OpenClaw framework development from this week and summarize the main technical highlights."
2. "Search for the current pros and cons of decentralized infrastructure for AI agents and provide a list of three reputable sources."
3. "Find information on the benefits of privacy-first search engines compared to traditional centralized alternatives."

Tips & Limitations

To maximize the utility of this skill, utilize the time parameter to filter results by day, week, month, or year when looking for time-sensitive data. Remember that the API is subject to a 100 requests/minute rate limit; if you are building an agent that handles heavy-duty scraping, implement an internal caching layer to store redundant results and prevent 429 error codes. Note that while the results are uncensored, the quality of the findings depends on the underlying index of the decentralized node network.