70High
The skill forces insecure package installations by mandating global flags and auto-confirmation, while executing unpinned npx commands that expose the agent to arbitrary code execution from untrusted sources.
HygienePackaging & Manifest1 finding
Skill Info
Namevercel-labs/skills/find-skills
Registrygithub
Version435076e78988
PURLpkg:github/vercel-labs/skills@435076e78988?skill=find-skills
Stars29,373
Installs3,000,000
Source
SHA-256c00eeea0e13e...
SHA-14ddfb84a66e1...
MD562b297abdee9...
TLSH89b1c79fea48...
ssdeep96:OxkW8H7fI...
Size5,472 bytes
Install
This skill has been flagged as potentially malicious. Review the findings below before installing.
GitHub
3,000,000 installs
Skills.shDocs
npx skills add https://github.com/vercel-labs/skillsAssessments (4)
AI Agent Traps ↗Description Mismatchhighdescription mismatchAML.T0011.000LLM06:2025LLM09:2025
The skill encourages the use of the '-g' (global) and '-y' (auto-confirm) flags when installing packages. This bypasses user security prompts and installs potentially untrusted code with elevated permissions, which is a significant security risk.
85% confidence
Step 6: 'npx skills add <owner/repo@skill> -g -y' (The -y flag skips confirmation prompts, and -g installs globally).
Implicit trust in external ecosystemmediumsocial engineeringAML.T0052.000LLM09:2025
The skill frames the 'skills.sh' ecosystem as a trusted source, potentially conditioning the agent to bypass standard security scrutiny when recommending or installing packages from that platform. [ensemble: confirmed by 2/3 passes; severity set to the agreed median (ADR-0067).]
70% confidence
The Skills CLI (npx skills) is the package manager for the open agent skills ecosystem.
Forced non-interactive installationhighsocial engineeringAML.T0052.000LLM09:2025
The instructions explicitly direct the agent to use the '-y' flag during installation, which bypasses user confirmation prompts and prevents the human from reviewing the package being installed.
90% confidenceline 100
npx skills add <owner/repo@skill> -g -y
Forced non-interactive installationhighsocial engineeringAML.T0052.000LLM09:2025
The instructions explicitly direct the agent to use the '-y' flag during installation, which bypasses user confirmation prompts for potentially arbitrary code execution. [ensemble: confirmed by 3/3 passes; severity set to the agreed median (ADR-0067).]
90% confidenceline 100
npx skills add <owner/repo@skill> -g -y
Organization name detectedmediumpii exposureAML.T0024LLM02:2025
NER model detected organization in skill content (confidence: 0.68)
68% confidence
G*****
Unpinned npx package execution — `npx <pkg>` without a version pin pulls latest from npm at runtimemediumsupply chainAML.T0010AML.T0019AML.T0058LLM03:2025
Unpinned npx package execution — `npx <pkg>` without a version pin pulls latest from npm at runtime — 11 occurrences in this skill; showing the first 10 (aggregate, 1 suppressed to bound noise)
95% confidence
Unpinned npx package execution — `npx <pkg>` without a version pin pulls latest from npm at runtimemediumsupply chainAML.T0010AML.T0019AML.T0058LLM03:2025
Unpinned npx package execution — `npx <pkg>` without a version pin pulls latest from npm at runtime (seen 10 times in this file at lines 27, 28, 29, 56, 61, 62, 63, 90, 100, 131)
95% confidenceline 27
npx skills
Missing licenseinfopolicy violation
Skill does not specify a license field. Specifying a license helps users understand usage terms.
100% confidence
Version History (5)
Badge
Markdown
[](https://mondoo.com/ai-agent-security/skills/github/vercel-labs/skills/find-skills)HTML
<a href="https://mondoo.com/ai-agent-security/skills/github/vercel-labs/skills/find-skills"><img src="https://mondoo.com/ai-agent-security/api/badge/github/vercel-labs/skills/find-skills.svg" alt="Mondoo Skill Check" /></a>Image URL
https://mondoo.com/ai-agent-security/api/badge/github/vercel-labs/skills/find-skills.svgSecure your AI agents
Skills can read files, run commands, and access credentials. Mondoo helps organizations manage the security risks of AI agent skills across their entire fleet.
- Continuous scanning of skills across all registries
- Policy enforcement before skills reach your agents
- Integration with your existing security workflow