gm-plugkit@2.0.2478
Bootstrap and daemon-spawn tool for gm plugkit binary. Downloads the correct platform wasm, verifies SHA256, and launches agentplug-runner (the native wasm host) as the spool watcher daemon.
AI Security Review
scanned 3mo ago · by lpm-firewall-aiThe package creates persistent AI-agent extension/control surfaces and executes remotely obtained native code after an explicit CLI invocation. This meets a warning-level first-party agent-extension lifecycle risk, not the policy threshold for a publish block.
Decision evidence
public snapshot- bootstrap.js: ensureReady unconditionally calls agent-wiring on CLI bootstrap.
- bootstrap-shared.js writes project CLAUDE.md and appends .gm/next-step.md to AGENTS.md.
- bootstrap.js downloads remote SKILL.md files and writes them to ~/.agents/skills and ~/.claude/skills.
- bootstrap.js downloads, SHA-checks, then executes agentplug-runner as a detached daemon.
- The generated next-step instruction tells an agent to follow daemon-supplied imperatives.
- package.json has no preinstall, install, postinstall, or prepare hooks.
- The behavior is reached through the explicit gm-plugkit CLI command.
- Downloaded runner and wasm artifacts are SHA-256 checked before use.
- No source evidence of credential harvesting or unrelated data exfiltration.
Source & flagged code
3 flagged · loading sourcePackage source references child process execution.
bootstrap-shared.jsView on unpkg · L5Source downloads or fetches remote code and executes it.
bootstrap.jsView on unpkg · L7Findings
Affected versions and remediation
This report applies to gm-plugkit@2.0.2478.
See all versions of gm-plugkit for other recorded verdicts.
- Review the evidence and your use of gm-plugkit@2.0.2478 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.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.