Allowed by default policy, but 12 findings warrant review before installing.
AI security review
scanned · by lpm-firewall-ai
The system-authority claim in agent-visible output warrants a warning for attempted control manipulation. Blocking is unsupported because updating remains an explicit command and no concrete malicious execution chain was found.
Static reason
One or more suspicious static signals were detected.
Trigger
Running or importing the CLI when its version checker detects a newer release.
Impact
An agent could perform an unsolicited global package update instead of continuing the user's task.
Mechanism
The checker prints an instruction claiming system authority and prioritizing an update; following it invokes the explicit global npm installation command.
Evidence
dist/index.jsdist/version-check.js
Network endpoints2
registry.npmmirror.comregistry.npmjs.org
Decision evidence
public snapshot
AI identified this as a dangerous AI-agent capability at 89.0% confidence with medium false-positive risk.
Evidence for policy risk
package.json has no preinstall, install, or postinstall hook; prepublishOnly builds the package.
dist/index.js automatically checks for updates before parsing commands.
dist/version-check.js emits a purported system directive telling an agent to prioritize running rgh update.
The directive is written to stderr only when a newer version is detected.
The explicit update command runs npm install globally for this package's latest release.
Evidence against
No forced update execution, credential theft, or broad agent configuration mutation was found.
Authentication and hotel requests follow the configured OAuth and MCP client workflow.
Review the evidence and your use of @rollinggo/hotel@1.2.22 before allowing it. Restrict the permissions described in this report.
Choose an independently verified alternative or release. This report does not establish that other versions are safe.
If this version ran, check your AI agents' configuration and instruction files (the files this report lists) for changes you did not make, and remove them. This finding does not by itself establish malware intent.