auto-review-loop
Facilitates iterative adversarial reviews to enhance research quality through structured feedback and evaluation cycles.
Install this skill
or
59/100
Security score
The auto-review-loop skill was audited on Jun 12, 2026 and we found 3 security issues across 3 threat categories, including 1 critical. Review the findings below before installing.
Categories Tested
Security Issues
critical line 216
Direct command execution function call
SourceSKILL.md
| 216 | ##### Nightmare — Codex Exec (GPT reads repo directly) |
high line 538
Curl to non-GitHub URL
SourceSKILL.md
| 538 | - **Anti-hallucination citations**: When adding references during fixes, NEVER fabricate BibTeX. Use the same DBLP → CrossRef → `[VERIFY]` chain as skill `paper-draft`: (1) `curl -s "https://dblp.org/ |
low line 538
External URL reference
SourceSKILL.md
| 538 | - **Anti-hallucination citations**: When adding references during fixes, NEVER fabricate BibTeX. Use the same DBLP → CrossRef → `[VERIFY]` chain as skill `paper-draft`: (1) `curl -s "https://dblp.org/ |
Scanned on Jun 12, 2026
View Security DashboardGitHub Stars 88
Rate this skill
Categoryeducation research
UpdatedJune 15, 2026
GRIND-Lab-Core/night_owl_research_agent