npkill
Verdict safe, but 1 finding (1 medium), mainly resource abuse. Most recent scan 2026-02-06.
The npkill skill is a legitimate utility for cleaning up Node.js build artifacts, but has moderate security and documentation concerns. The primary issues are: (1) lack of explicit tool declarations despite heavy bash command usage, (2) documentation of a potentially dangerous automated deletion command without enforcement of safety checks, (3) confusing references to non-existent Python files, and (4) missing optional metadata. The skill does not contain malicious code, data exfiltration, or prompt injection attempts. However, the automated deletion capability and lack of tool restrictions present moderate risks if the agent executes commands without proper user oversight. The skill would benefit from explicit tool declarations, removal of the automated deletion example or stronger safety enforcement, and cleanup of erroneous file references.
Scanner's overall assessment (LLM analyzer)
Second scanner: NVIDIA SkillSpector
Independent static run, no language model, over the same files as the Cisco scan above. How the two scanners compare across the whole corpus: Two scanners, one corpus.
What this skill does
Clean up node_modules and .next folders to free up disk space using npkill. Specifically designed to help JavaScript and Next.js developers remove accumulated build artifacts that consume significant storage. Provides both interactive and automated cleanup options with safety checks to protect important system directories.
- Package
- 1 file, 3.7 KB
- SKILL.md
- 505 words, 12+ sections
- Code
- None. Instructions only.
SKILL.md covers: NPkill - Node.js and Next.js Build Artifact Cleaner · Purpose · When to Use This Skill · Core Commands · Interactive Cleanup (Recommended) · Target .next Folders Specifically · Dry Run (Always Recommended First) · Automated Cleanup (Use with Caution) · View Sizes in Gigabytes · Scan from Specific Directory · Safety Features · Common Use Cases for Next.js Developers
Findings
MEDIUM Potentially Dangerous Automated Deletion Command Resource Abusellm
The skill documents an automated deletion command 'npkill --delete-all --yes' that bypasses all confirmations and deletes all found node_modules folders without user review. While the instructions include a warning 'Use with Caution', the agent could execute this command if instructed, potentially causing unintended data loss. The skill encourages using dry-run first, but there's no enforcement mechanism to prevent immediate execution of the destructive command.
SKILL.md
### Automated Cleanup (Use with Caution)
```bash
npkill --delete-all --yes
```
Automatically deletes all node_modules folders found. Use only after verifying with dry-run.What is in the package
| File | Kind | Size |
|---|---|---|
SKILL.md | doc | 3.7 KB |
Package contents recorded 2026-10-22. The scan above ran against the version current at scan time; the two can differ if the skill has since been updated.
What we checked
- Static analyzer. Pattern rules over SKILL.md and bundled files: known-bad commands, exfiltration URLs, secrets, obfuscation.
- Behavioral analyzer. Checks what the skill asks the agent to do against what its description claims it does.
- Llm analyzer. A language model reads the skill and reports intent-level issues the rules cannot express.
- Trigger analyzer. Looks for instructions that fire on conditions the user did not ask for.
- Meta analyzer. A second model pass that validates, deduplicates and re-scores the other analyzers' findings.
What we did not check
- Runtime behaviour. The skill was not executed; findings come from reading its files.
- Anything fetched at install or run time, including remote scripts, package dependencies, and URLs the skill tells the agent to open.
- Content the analyzers cannot read: images, binaries, and some non-English text.
- Known scanner bypass classes. See the methodology and limitations page.
Sources
- Repository: https://clawhub.ai/skills/npkill
- Raw scan result: npkill-scan.json
- Badge for your README:
- Request a re-scan