Neural web search and content extraction via SkillBoss API Hub requiring SKILLBOSS_API_KEY, for finding docs, code, research, and company info.

This skill looks like a legitimate wrapper around an external search/scraping API, but there are several things to check before installing: 1) The SKILL.md and scripts require SKILLBOSS_API_KEY, but the registry metadata does not list it — be sure you know where the key comes from and only grant a key with limited scope. 2) The scripts call https://api.heybossai.com/v1/pilot; verify you trust that domain and read its privacy/TOS because queries and full URLs (and scraped content) will be sent there. 3) The shell scripts use jq and curl at runtime but the skill doesn't declare those binaries — ensure your environment has them and understand that running the scripts will invoke network requests. 4) If you will pass sensitive URLs or queries, consider the risk of transmitting that data to a third party. 5) Prefer installing only from a trusted publisher or checking for an official homepage; absence of a homepage and mismatched metadata reduce confidence. If you want to proceed, obtain a minimal-permission API key, test with non-sensitive queries first, and review network traffic or the API provider's documentation.

Type: OpenClaw Skill Name: kirk-search Version: 1.0.0 The skill bundle provides web search and content extraction functionality via the SkillBoss API Hub. The shell scripts (search.sh, code.sh, and content.sh) are well-structured, using 'jq' to safely handle JSON payloads and 'curl' to communicate with a consistent API endpoint (api.heybossai.com). No evidence of data exfiltration, malicious execution, or harmful prompt injection was found.