registry / @wrongstack/core / 0.282.0

@wrongstack/core@0.282.0

WrongStack core: kernel, types, defaults, and shared utilities for the WrongStack CLI agent.

AI Security Review

scanned 10d ago · by lpm-firewall-ai

No confirmed malicious attack surface from static inspection. The package exposes agent framework utilities with user-invoked network, filesystem, child-process, MCP, and skill-install capabilities aligned with WrongStack.

Static reason

No blocking static signals were detected.

Trigger

Importing library modules or explicitly invoking WrongStack tools/skills at runtime

Impact

No evidence of credential harvesting, exfiltration, persistence, destructive install behavior, or unconsented AI-agent control-surface mutation.

Mechanism

Framework utilities and explicit user-command tools, not lifecycle execution

Rationale

Source inspection shows a broad agent framework with dangerous primitives gated behind explicit runtime APIs and permission metadata, with no npm lifecycle hooks or import-time attack chain. Network and environment references are package-aligned and user-invoked rather than stealth exfiltration.

Evidence

package.jsonREADME.mddist/index.jsdist/utils/index.jsdist/tools/index.jsdist/skills/index.jsdist/security/index.jsskills/wrongstack-mailbox/SKILL.md~/.wrongstack/<project>/.wrongstack/AGENTS.md<project>/.wrongstack/skills/~/.claude/skills/

Network endpoints6

skills.sh

@wrongstack/core@0.282.0

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Source & flagged code