ClawHub
Back to skill

Security audit

nvimclaw

Security checks across malware telemetry and agentic risk

Overview

The skill is a disclosed Neovim bridge that lets an agent read and, with explicit privileged setup, edit live Neovim buffers within the configured workspace.

Install this only if you want an agent connected to your live Neovim session. Keep the safe tier unless you need edits, use explicit command allowlists on shared gateways, avoid enabling broad nvim.* privileged access unless you trust the setup, and remember that chat sends can include current-buffer context unless disabled.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Vague Triggers

Medium
Confidence
92% confidence
Finding
The skill recommends using nvimclaw for very broad categories of requests like reading, editing, or inspecting anything in a live Neovim instance. That broad activation scope can cause an agent to invoke a high-capability editor bridge in situations where a lower-risk workflow would suffice, increasing the chance of unintended access to sensitive buffers or unintended stateful actions.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The introductory guidance presents the bridge as a general-purpose way to read and edit a live editor session without an upfront warning that some available commands are privileged and can mutate buffers, cursor position, or UI state. This can normalize use of a powerful live-edit channel before the user is clearly informed of the risk of data loss, disruption, or unintended modifications.

Session Persistence

Medium
Category
Rogue Agent
Content
Full-buffer overwrite. Alias for `replace_lines(0, -1, lines)` with the same conflict semantics. Provided for agents trained on `vscode.file.write`; **prefer `nvim.ex.substitute` or `nvim.buffer.replace_lines` when possible** — they preserve Vim's undo history per edit.

```bash
openclaw nodes invoke --node <N> --command nvim.buffer.write \
  --params '{
    "path": "drafts/example.md",
    "content": "Schopenhauer is hilarious. ...",
Confidence
60% confidence
Finding
Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.