sundial-org/project-referee
skills/project-referee/SKILL.md
Critiques ML conference papers with reviewer-style feedback. Use when users want to anticipate reviewer concerns, identify weaknesses, check claim-evidence gaps, or find missing citations.
$ install --global
skillsauthnpx skillsauth add sundial-org/skills project-refereeInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: Apr 15, 2026, 5:08 AM37.4s3 files scanned
SKILL.md
- name:
- project-referee
- description:
- Critiques ML conference papers with reviewer-style feedback. Use when users want to anticipate reviewer concerns, identify weaknesses, check claim-evidence gaps, or find missing citations.
Project Referee
Review Workflow
Step 1: Assess Manuscript Stage
Ask or determine the manuscript stage:
- Early draft: Focus on novelty, feasibility, relevance, structure
- Mid-stage draft: Focus on development trajectory, completeness, preliminary results
- Final submission: Apply full review criteria (see references/reviewer-instructions.md)
Step 2: Stage-Specific Review
Early Draft Protocol:
- Assess core contribution novelty and significance
- Evaluate method feasibility with current technology
- Check alignment with venue scope
- Review logical structure and completeness
- Identify major missing components
- Suggest high-level direction
Mid-Stage Protocol:
- Evaluate development progress
- Check section completeness
- Assess preliminary results against claims
- Review literature coverage
- Identify specific weaknesses
- Provide targeted refinements
Final Submission Protocol: Apply complete review process. See references/reviewer-instructions.md for the full reviewer form.
Step 3: Search for Related Work
Proactively search for missing citations:
- Recent work in the paper's domain
- Papers with similar methods or contributions
- Relevant benchmarks and baselines
- Use WebSearch to find recent publications
Step 4: Provide Structured Feedback
Use the output format in references/reviewer-instructions.md:
- Summary, Strengths, Weaknesses, Questions
- Soundness (1-4), Presentation (1-4), Contribution (1-4)
- Overall Score (1-10), Confidence (1-5)
Neuro-Symbolic Papers
For papers combining LLMs with symbolic reasoning, see references/neuro-symbolic-review-criteria.md for additional criteria:
- Symbolic formulation quality
- Solver selection and integration
- Faithfulness and verification
- Comparison to baselines (Logic-LM, LINC)
References
- references/reviewer-instructions.md - Full reviewer form
- references/neuro-symbolic-review-criteria.md - Neuro-symbolic specific criteria
Related Skills
sundial-org/cs448b-visualization
development
VerifiedTrustedCommunity
Data visualization design based on Stanford CS448B. Use for: (1) choosing chart types, (2) selecting visual encodings, (3) critiquing visualizations, (4) building D3.js visualizations, (5) designing interactions/animations, (6) choosing colors, (7) visualizing networks, (8) visualizing text. Covers Bertin, Mackinlay, Cleveland & McGill.
148SKILL.mdUpdated Apr 15, 2026
sundial-org/training-data-curation
testing
VerifiedTrustedCommunity
Guidelines for creating high-quality datasets for LLM post-training (SFT/DPO/RLHF). Use when preparing data for fine-tuning, evaluating data quality, or designing data collection strategies.
148SKILL.mdUpdated Apr 15, 2026
sundial-org/tinker
development
VerifiedTrustedCommunity
Fine-tune LLMs using the Tinker API. Covers supervised fine-tuning, reinforcement learning, LoRA training, vision-language models, and both high-level Cookbook patterns and low-level API usage.
148SKILL.mdUpdated Apr 15, 2026
sundial-org/tinker-training-cost
data-ai
VerifiedTrustedCommunity
Calculate training costs for Tinker fine-tuning jobs. Use when estimating costs for Tinker LLM training, counting tokens in datasets, or comparing Tinker model training prices. Tokenizes datasets using the correct model tokenizer and provides accurate cost estimates.
148SKILL.mdUpdated Apr 15, 2026