thesis-excellence
Skill17 starsChanged 9 months agoMITView on GitHub
Multi-agent thesis review — dispatches specialized agents based on scope, synthesizes combined report
In this repository11 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- build-twfe-explainerSkill
- check-methodologySkill
- compile-thesisSkill
- generate-mapsSkill
- proofreadSkill
- review-codeSkill
- rewrite-literatureSkill
- verify-claimsSkill
- write-introductionSkill
- write-sectionSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 20
Run tectonic to verify compilation:
- Distinctive: line 24
If compilation fails, report the error and stop.
- Distinctive: line 40
Use the Task tool to fork to each agent in parallel where possible.
- Distinctive: line 41
proofreader: Read the chapter(s) and produce a proofreading report
- Distinctive: line 43
notation-checker: Check all math against the notation registry
- Distinctive: line 91
If an agent fails or times out, note it and continue with others
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Thesis Excellence Review
Orchestrate a comprehensive multi-agent review of the thesis. The argument specifies the scope.
Scope Options
all— Review the entire thesis (all written chapters)04_methodology/05_results/ etc. — Review a specific chaptercode— Review all analysis notebookspre-submit— Full pre-submission check (everything)
Orchestration Procedure
Step 1: Compile
Run tectonic to verify compilation:
cd /Users/alessandro/Projects/Tesi/thesis && tectonic main.tex 2>&1If compilation fails, report the error and stop.
Step 2: Determine Agents to Dispatch
Based on the scope, select which agents to run:
| Scope | Agents |
Any .tex chapter | proofreader, latex-auditor, notation-checker |
| Ch. 4 or Ch. 5 | + causal-inference |
Chapters with \citet/\citep | + literature-rag |
code scope | code-reviewer |
all or pre-submit | All agents |
Step 3: Run Agents
Use the Task tool to fork to each agent in parallel where possible. For each agent:
- proofreader: Read the chapter(s) and produce a proofreading report
- latex-auditor: Compile and check cross-references, citations, formatting
- notation-checker: Check all math against the notation registry
- causal-inference: (only for Ch. 4/5) Deep methodology review
- literature-rag: (only for chapters with citations) Verify top claims via RAG
- code-reviewer: (only for code scope) Review notebooks
Step 4: Synthesize Report
Combine all agent reports into a unified assessment:
THESIS EXCELLENCE REPORT
=========================
Scope: [what was reviewed]
Date: [date]
COMPILATION: [PASS/FAIL]
AGENT REPORTS:
├── Proofreader: [summary + issue count]
├── LaTeX Auditor: [summary + issue count]
├── Notation Checker: [summary + issue count]
├── Causal Inference: [summary + top recommendation]
├── Literature RAG: [summary + verification rate]
└── Code Reviewer: [summary + issue count]
QUALITY ASSESSMENT:
├── Clarity: X/10
├── Rigor: X/10
├── Consistency: X/10
├── Completeness: X/10
└── Presentation: X/10
Overall: X.X/10
TOP 5 PRIORITY ACTIONS:
1. [highest priority improvement]
2. [second priority]
3. [third priority]
4. [fourth priority]
5. [fifth priority]
DETAILED FINDINGS:
[Include full agent reports below, organized by agent]Important
- Quality scores are ADVISORY ONLY — they never block any action
- Focus the "priority actions" on items that would most improve the thesis
- If an agent fails or times out, note it and continue with others
- For
pre-submitscope, be especially thorough — this is the final check before submission
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

