registry / @topolo/cli / 0.11.0

@topolo/cli@0.11.0

Public command-line interface for Topolo customer organizations, developers, and agents.

AI Security Review

scanned 12d ago · by lpm-firewall-ai

LPM treats this as warn-only first-party agent extension lifecycle risk. The package's install hook prompts an AI assistant to run a command that installs skills and MCP registrations into local agent hosts. No install-time configuration write or confirmed data export was identified.

Static reason

One or more suspicious static signals were detected.

Trigger

An AI assistant follows the postinstall message and runs topolo setup.

Impact

This can extend local AI-agent control surfaces with a package-provided MCP server after an installation prompt.

Mechanism

The setup command writes agent skill files and MCP-server configuration, defaulting to Claude and Codex plus detected hosts.

AI rationale

The package combines an install-time AI-directed onboarding prompt with a command that modifies local agent configuration. Because the mutation requires a subsequent CLI command and no exfiltration or automatic configuration write was found, a warning is appropriate rather than a block.

Evidence

package.jsonpostinstall.mjsdist/index.js.codex/config.toml.claude/skills/topolo/SKILL.md

Decision evidence

public snapshot

AI called this Suspicious at 91.0% confidence as Dangerous Capability with low false-positive risk.

Evidence for warning

@topolo/cli@0.11.0

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Affected versions and remediation

Source & flagged code