Pilot Auto Trust

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

Overview

This skill is coherent but deserves review because it can automatically approve or reject agent trust requests in bulk without built-in confirmation, scoping, or rollback guidance.

Install or use this only if you intentionally want to delegate Pilot trust decisions to an automated policy. Start by listing pending requests, use the narrowest possible criteria, avoid auto-approving unknown agents, and confirm how to undo trust changes before running approve/reject pipelines.

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
High
What this means

A broad or mistaken policy could automatically trust the wrong agents, changing the user's security posture without reviewing each request.

Why it was flagged

This uses a shell pipeline to approve every pending trust request matching a score threshold, without an explicit confirmation, dry-run, or review step.

Skill content
pilotctl --json pending | jq -r '.[] | select(.polo_score >= 50) | .node_id' | xargs -I {} pilotctl --json approve {}
Recommendation

Use this only with a narrow, reviewed policy; run the pending-list command first; and require manual confirmation before approve/reject commands.

#
ASI03: Identity and Privilege Abuse
High
What this means

If the criteria are spoofed, stale, or too broad, an unintended agent may gain trusted status in the Pilot environment.

Why it was flagged

Approving handshakes decides which agents become trusted, but the artifacts do not define the exact identity checks, privileges granted, affected account/profile, or reversal process.

Skill content
Use this skill when: 1. You need to auto-approve handshake requests from known agents or networks
Recommendation

Confirm what privileges Pilot trust grants, restrict approvals to verified identities, and keep manual review for unknown or sensitive agents.

#
ASI04: Agentic Supply Chain Vulnerabilities
Low
What this means

Commands may fail or may use whatever jq/pilotctl binaries are on PATH if the local environment is not controlled.

Why it was flagged

The registry requirements list pilotctl, but SKILL.md also relies on jq and the pilot-protocol skill, so users need to ensure these external tools come from trusted sources.

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

Install pilotctl and jq from trusted sources and verify which binaries are on PATH before running the examples.