Pilot Clipboard

Shared clipboard for quick text and data snippets between agents over Pilot Protocol. Use this skill when: 1. You need to share short text snippets or comman...

This skill appears to do what it says (share short text over the Pilot network), but take these precautions before installing or using it: - Do not send secrets: Clipboard contents and command output sent with pilot-clipboard will be transmitted to remote Pilot peers. Avoid copying passwords, API keys, or anything sensitive. - Verify the pilotctl binary and Pilot network: pilotctl is the program actually performing network operations. Only use pilot-clipboard if pilotctl comes from a trusted source and you understand how Pilot identifies recipients. Check the pilot-protocol skill/daemon configuration for where credentials are stored and who can receive messages. - Install missing tools: The SKILL.md examples require jq and (optionally) pbcopy/pbpaste or xclip/xsel for system clipboard integration; these are not declared in metadata. Install these separately if you intend to use the examples. - Test in isolation: Try the workflow in a sandbox or test account before using it with real data. - Confirm provenance: The skill source is unknown and the registry entry points to a homepage; if you need higher assurance, ask the publisher for source code or a signed release and review the pilotctl/pilot-protocol codebase. Confidence is medium: the skill is coherent with its stated purpose, but the undeclared runtime dependencies and the inherent privacy/exfiltration risk of a networked clipboard justify caution.

Type: OpenClaw Skill Name: pilot-clipboard Version: 1.0.0 The pilot-clipboard skill facilitates remote data transfer and inter-agent communication using the pilotctl binary and a custom 'Pilot Protocol.' While the functionality is aligned with its stated purpose of a shared clipboard, it provides a high-risk mechanism for data exfiltration to arbitrary remote addresses (e.g., '1:0001.AAAA.BBBB'). The examples in SKILL.md demonstrate sending command outputs and arbitrary data strings to external endpoints, which could be abused to leak sensitive information if the agent is compromised or misdirected.