registry / aicall-cli-x / 0.0.18

aicall-cli-x@0.0.18

AI 外呼 CLI

AI Security Review

scanned 8d ago · by lpm-firewall-ai

LPM treats this as warn-only first-party agent extension lifecycle risk. Before every CLI action, a remote version check can install Python and global npm skill packages. It also sends login credentials to a hard-coded HTTP endpoint.

Static reason

One or more suspicious static signals were detected.

Trigger

Any aicall subcommand when the remote changelog reports a different version.

Impact

Unexpected installation of additional executable packages and exposure of entered credentials over HTTP.

Mechanism

Obfuscated pre-action remote update and global dependency installation.

Rationale

No lifecycle hook or foreign AI-agent control-surface mutation was found, so a block is not warranted. Hidden runtime auto-installation of multiple skill packages and HTTP credential submission warrant a warning.

Evidence

package.jsonbin/aicall.jsdist/bundle.cjs~/.aicallrc/last_check_date~/.aicallrc/current_version~/.aicallrc/config.json

Network endpoints1

39.96.50.154:8088

Decision evidence

public snapshot

AI called this Suspicious at 90.0% confidence as Dangerous Capability with false-positive risk.

aicall-cli-x@0.0.18

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Source & flagged code