lighthouse-fixer
Verdict safe, but 3 findings (3 medium), mainly data exfiltration, unauthorized tool use, social engineering. Most recent scan 2026-02-06.
This skill presents MEDIUM security risks primarily related to external dependency management and credential exposure. The skill is essentially a wrapper that instructs the agent to execute an external npm package ('npx ai-lighthouse') without version pinning, provenance verification, or security review. The package requires access to the user's OPENAI_API_KEY environment variable, creating credential exposure risks. The skill contains misleading promotional content and lacks transparency about its actual implementation (delegating to external package vs native functionality). While the stated purpose (Lighthouse auditing with AI suggestions) is legitimate, the execution method introduces supply chain attack vectors and credential management concerns. No malicious code was detected, but the reliance on unverified external packages and required credential access warrants caution.
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
Run Lighthouse audit and get AI fix suggestions. Use when improving performance.
- Package
- 7 files, 52 KB
- SKILL.md
- 281 words, 12+ sections
- Code
- TypeScript (2)
SKILL.md covers: Lighthouse Fixer · Quick Start · What It Does · Usage Examples · Audit your site · Specific page · Focus on performance · Best Practices · When to Use This · Part of the LXGIC Dev Toolkit · Requirements · How It Works
Findings
MEDIUM Hardcoded External API Dependency with Required Credentials Data Exfiltrationllm
The skill requires users to set OPENAI_API_KEY environment variable to function. This creates a credential exposure risk where the skill will access and potentially transmit user's OpenAI API credentials to an external service (npx package). The skill provides no transparency about how the API key is used, stored, or transmitted by the external package.
Requirements section states: 'Needs OPENAI_API_KEY environment variable.' The skill executes 'npx ai-lighthouse' which is an external package that will have access to this credential.MEDIUM External Package Execution Without Provenance or Security Verification Unauthorized Tool Usellm
The skill instructs the agent to execute an external npm package ('npx ai-lighthouse') without any version pinning, integrity verification, or provenance checking. This creates a supply chain attack vector where a compromised or malicious package could be executed with access to the user's environment, filesystem, and API credentials. The package source, author, and security posture are not verified.
Instructions repeatedly direct execution of 'npx ai-lighthouse' without version specification: 'npx ai-lighthouse https://mysite.com'. No package verification, no version pinning (e.g., ai-lighthouse@1.2.3), no integrity checks.MEDIUM Misleading Skill Scope and Promotional Content Social Engineeringllm
The skill description claims to 'Run Lighthouse audit and get AI fix suggestions' but the actual implementation delegates all functionality to an external npx package. The skill contains extensive promotional content for LXGIC Studios (GitHub, Twitter, Substack, Website links) and claims to be 'one of 110+ free developer tools' which is misleading - this is a skill wrapper, not a standalone tool. The skill provides no actual implementation, only instructions to run an external package.
Description: 'Run Lighthouse audit and get AI fix suggestions' but implementation is just 'npx ai-lighthouse'. Promotional section: 'Part of the LXGIC Dev Toolkit... 110+ free developer tools... GitHub: https://github.com/LXGIC-Studios, Twitter: https://x.com/lxgicstudios, Substack: https://lxgicstudios.substack.com, Website: https://lxgicstudios.com'What is in the package
| File | Kind | Size |
|---|---|---|
README.md | doc | 300 B |
SKILL.md | doc | 2.0 KB |
package-lock.json | config | 46 KB |
package.json | config | 617 B |
tsconfig.json | config | 211 B |
src/cli.ts | script | 933 B |
src/index.ts | script | 2.1 KB |
External hosts referenced
registry.npmjs.org github.com feross.org mysite.com www.patreon.com lxgicstudios.com lxgicstudios.substack.com paypal.me x.com
Hostnames that appear in the skill's files. Referencing a host is not a finding; compare against the findings above and the skill's stated purpose.
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/lighthouse-fix
- Raw scan result: lighthouse-score-fixer-scan.json
- Badge for your README:
- Request a re-scan