Pilot Trust Circle

Security checks across static analysis, malware telemetry, and agentic risk

Overview

This instruction-only skill is coherent, but it automates persistent trust approvals between Pilot agents, so users should verify circle membership before using it.

Install only if you intend to automate Pilot agent trust circles. Before running bootstrap or add-member workflows, review the circle JSON files, confirm each agent identity, and avoid using this skill where you need per-connection manual approval.

Static analysis

No static analysis findings were reported for this release.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

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.

#
ASI02: Tool Misuse and Exploitation
Medium
What this means

Running the documented workflows can grant trust to Pilot agents, which may affect what those agents can do or exchange with each other.

Why it was flagged

The skill uses Bash and pilotctl to initiate handshakes and approve nodes. This is aligned with the trust-circle purpose, but it changes agent trust relationships.

Skill content
pilotctl --json handshake "$NEW_MEMBER" "Member of $CIRCLE"
NODE_ID=$(pilotctl --json find "$NEW_MEMBER" | jq -r '.[0].node_id')
pilotctl --json approve "$NODE_ID"
Recommendation

Only run these workflows for circles and members you intend to trust, and consider reviewing each node identifier before approving it.

#
ASI07: Insecure Inter-Agent Communication
Medium
What this means

If an incorrect or unintended agent is included in a circle, it may become trusted by other agents in that group.

Why it was flagged

The skill intentionally creates inter-agent trust relationships. That is disclosed and purpose-aligned, but users should understand the identity and data-sharing implications of mutual trust.

Skill content
Manage named trust groups where all members automatically trust each other.
Recommendation

Verify agent names and node identities before bootstrapping a circle, and use manual trust approval when you need fine-grained or hierarchical trust.

#
ASI06: Memory and Context Poisoning
Low
What this means

Stale or modified circle files could cause later workflows to trust agents the user no longer intends to trust.

Why it was flagged

The skill stores circle membership in persistent local JSON files that are later read to decide which agents to handshake with or approve.

Skill content
mkdir -p ~/.pilot/circles
cat > ~/.pilot/circles/team-alpha.json <<EOF
Recommendation

Keep ~/.pilot/circles files protected and review them before running bootstrap or approval workflows.

#
ASI04: Agentic Supply Chain Vulnerabilities
Info
What this means

The skill may fail or behave unexpectedly on systems without jq installed.

Why it was flagged

The SKILL.md documents jq as a dependency, while the registry requirements list only pilotctl. This is a metadata completeness issue rather than evidence of malicious behavior.

Skill content
Requires pilot-protocol skill, pilotctl, and jq.
Recommendation

Ensure jq is installed before use, and the publisher should declare it in the required binaries metadata.