Scan Report
20 /100
gitlab-auto-review
Automated AI code review for GitLab Merge Requests via polling
Legitimate GitLab MR code review automation skill with minor documentation gaps but no malicious behavior detected.
Safe to install
Add explicit filesystem:READ declaration in SKILL.md for the --file mode JSON reading. No action needed to block this skill.
Findings 2 items
| Severity | Finding | Location |
|---|---|---|
| Low | Undeclared filesystem read operation | scripts/gitlab-api.js:138 |
| Info | No dependency pinning | SKILL.md:12 |
| Resource | Declared | Inferred | Status | Evidence |
|---|---|---|---|---|
| Network | READ | READ | ✓ Aligned | scripts/gitlab-api.js:27 Makes GET/POST requests to GitLab API only |
| Shell | WRITE | WRITE | ✓ Aligned | SKILL.md:18-22 Declares node command execution; cron-setup.md:7 documents full c… |
| Environment | READ | READ | ✓ Aligned | scripts/gitlab-api.js:14-15 Reads GITLAB_URL and GITLAB_TOKEN from process.env |
| Filesystem | NONE | READ | ✓ Aligned | scripts/gitlab-api.js:138 uses fs.readFileSync for --file mode; should be declar… |
| Database | NONE | NONE | — | No database operations found |
| Browser | NONE | NONE | — | No browser operations found |
| Skill Invoke | NONE | NONE | — | No nested skill invocations |
| Clipboard | NONE | NONE | — | No clipboard operations found |
File Tree
4 files · 12.2 KB · 357 lines Markdown 3f · 184L
JavaScript 1f · 173L
├─
▾
references
│ ├─
cron-setup.md
Markdown
│ └─
review-guidelines.md
Markdown
├─
▾
scripts
│ └─
gitlab-api.js
JavaScript
└─
SKILL.md
Markdown
Dependencies 1 items
| Package | Version | Source | Known Vulns | Notes |
|---|---|---|---|---|
node | unspecified | system | No | No version constraint; assumes Node.js runtime is available |
Security Positives
✓ All network requests are to GitLab API only (declared intent matches implementation)
✓ No credential exfiltration or suspicious data transmission
✓ No base64-encoded payloads, obfuscation, or hidden instructions
✓ No access to sensitive paths (~/.ssh, ~/.aws, .env credentials)
✓ Clean codebase with no eval(), exec(), or shell injection vectors
✓ Proper JSON parsing with error handling for API responses
✓ Follows security best practices for comment posting via --file mode (Windows encoding fix)
✓ No curl|bash or remote script execution patterns