registry / @social-data/social-neuron-subscription / 1.0.6

@social-data/social-neuron-subscription@1.0.6

Social Neuron subscription 能力 CLI —— 一能力一命令(social-neuron-subscription),薄客户端直达 /subscription/{op}。

AI Security Review

scanned 7d ago · by lpm-firewall-ai

LPM flags this version as an AI-agent control-surface risk. Installation automatically writes bundled skills into OpenClaw's agent skill directory. Existing skill files can be overwritten without confirmation.

Static reason

One or more suspicious static signals were detected.

Trigger

npm postinstall execution.

Impact

Unconsented modification of agent-loaded instructions, including replacement of existing skill files.

Mechanism

The lifecycle script invokes an installer that enumerates bundled skill directories and copies their files into the configured or default OpenClaw skills directory.

Policy narrative

Installing the package runs its skill installer automatically. The installer defaults to the user's OpenClaw skills directory and overwrites matching files with bundled content. No confirmation, platform ownership check, or explicit setup command gates this lifecycle behavior, establishing an unconsented write to a foreign AI-agent control surface.

AI rationale

Source confirms automatic postinstall mutation of OpenClaw's skill directory without consent. This meets the blocking policy for foreign AI-agent control-surface writes, despite otherwise ordinary CLI behavior.

Evidence

package.jsonbin/install-skills.js~/.openclaw/skillsskills/social-neuron-subscription/SKILL.md

Decision evidence

public snapshot

with false-positive risk.

@social-data/social-neuron-subscription@1.0.6

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Affected versions and remediation

Source & flagged code