Pilot Swarm Join
Security checks across static analysis, malware telemetry, and agentic risk
Overview
The skill is coherent for joining agent swarms, but it includes an unfiltered command that approves every pending peer, which could trust unintended agents.
Review peer identities before using this skill. Avoid running the approve-all command as written unless you are certain every pending peer is intended to join the swarm.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
An unrelated or unexpected pending peer could become trusted and participate in agent-to-agent communication.
This command approves every pending node ID without filtering to the intended swarm, expected registry, discovered peer list, or an explicit user confirmation step.
pilotctl --json pending | jq -r '.[].node_id' | xargs -I {} pilotctl --json approve {}Require explicit user approval before trusting peers, show peer identities and swarm membership, and filter approvals to the intended swarm and registry.
The safety of this instruction-only skill depends on the behavior and configuration of the local pilotctl tooling and daemon.
The skill depends on external local tools and a running daemon; this is purpose-aligned, but users should ensure those components are installed from trusted sources and configured intentionally.
Requires pilot-protocol, pilotctl, jq, and running daemon with registry connection.
Verify the installed pilotctl and daemon source, and ensure jq and pilot-protocol components are expected before using the skill.