awesome-latex-skills
by Calix-L
LaTeX manuscript workflows with five agent skills, project diagnostics, reproducible builds and offline change review.
187stars
4forks
Python
Added 5/16/2026
From $49 · 30 days · terms & refunds
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
Maintainer of this skill?
Automated security scan — not a certification.
Installation
# Add to your Claude Code skills
git clone https://github.com/Calix-L/awesome-latex-skillsGetting Started
Guides for using ai agents skills like awesome-latex-skills.
- Caveman: Cut Claude Token Use by 65%How agent-side prompt compression works, when to use it, and when not to.
- How to Test an AI Skill: A Practical Guide to EvalsEvaluate skills with realistic fixtures, a baseline, and repeatable checks.
- Getting Started with AI SkillsFirst-time install walkthrough for Claude Code, Codex CLI, and ChatGPT.
Security ReportVerified
Last scanned: 5/30/2026
{
"issues": [],
"status": "PASSED",
"scannedAt": "2026-05-30T15:54:18.430Z",
"npmAuditRan": true,
"pipAuditRan": true
}Compare with similar skills
See how awesome-latex-skills compares with popular alternatives.
Skills / Setup / Examples / Docs / FAQ
English / 简体中文
01 / Choose your skill
Start with the problem in front of you. Each entry includes a workflow, references, and verification requirements.
| Skill | Purpose & deliverable |
|---|---|
| latex-rescue01 / REPAIR | Diagnose build errors, make minimal source fixes, and report build evidence and remaining issues. |
| latex-polish02 / POLISH | Refine academic prose while preserving claims and numbers, at light, moderate, or strict intensity. |
| latex-fmt03 / FORMAT | Migrate publication templates and report passed, failed, and unverified requirements against official guidance. |
| paper-read04 / READ | Connect claims to evidence, from a quick skim to deeper appraisal, with locations in the original paper. |
| pdf2tex05 / RECOVER | Reconstruct editable LaTeX from a PDF, flag uncertain content, and compare it with the original. |
NeurIPS · ICML · CVPR · ACL/EMNLP · ICLR · ECCV · AAAI · TMLR · IEEE · Nature · Science · COLING · KDD · SIGIR · Interspeech.
The venue guide routes to official instructions. Verify the requested year, track, article type, and submission stage; obtain official author kits separately.
02 / Get started
Python 3.10+ · Windows / macOS / Linux · No third-party installer dependencies
git clone https://github.com/Calix-L/awesome-latex-skills.git
cd awesome-latex-skills
python -m pip install .
als install --agent claude
Using Codex? Change the last command to als install --agent codex.
Use python3 if that is your system's Python command.
Package installation & optional dependencies. Prefer a
checkout without installation? python scripts/als.py accepts the same commands.
Invoke: /latex-rescue in Claude Code or $latex-rescue in Codex. Natural-language requests work too.
Get your first useful result. Follow the five task recipes for commands, expected outputs and the next check.
| Start here | What you can review |
|---|---|
| Copy an editable case | Six cases available from the installed wheel, with an offline guide and source fingerprints |
| Diagnose your project | Actual settings, local tools, missing resources and source locations |
| Inspect manuscript changes | Original/candidate differences, author decisions and retained build evidence |
| Try the complete case | A multi-file English/Chinese manuscript and an offline review report |
Start with your own copy; no TeX or PDF library is required for export:
als examples export --case full-paper --output ../my-example
als verify example ../my-example
Open ../my-example/report.html, then follow its commands. Preserve the
initial export and edit a separate candidate. All six cases & prerequisites.
Planning to assess skill quality? Start with the 12-run evaluation pilot. Read the tested environments and interface policy before integrating reports.
# Preview first; no files are written
python scripts/install.py --agent codex --skill latex-rescue --dry-run
# Install only the selected bundle; repeat --skill to select more
python scripts/install.py --agent codex --skill latex-rescue
# Choose your own skills directory
python scripts/install.py --dest "path/to/skills" --skill paper-read
Each installation includes the entrypoint, references, scripts, and agent metadata.
Default destinations are ~/.claude/skills/ for Claude Code and
$CODEX_HOME/skills/ for Codex (default ~/.codex/skills/).
Identical bundles stay untouched; ordinary installation never replaces differing content.
git pull --ff-only
python scripts/install.py --agent codex --update --dry-run
python scripts/install.py --agent codex --update
New installations record file SHA-256 hashes. --update replaces only bundles
that still match their receipt; local edits, additions, or deletions stop the
whole batch. There is no force-overwrite mode. Failed or interrupted updates
attempt to restore the old directories; failed rollback preserves recovery files
and reports their location.
Manually copied bundles without receipts can be adopted only when they exactly
match the current checkout (adopt). Back up and move differing old copies away,
or choose a new destination. --dry-run writes neither files nor receipts.
Use --agent claude for Claude Code.
Writers to the same skills directory use an exclusive lock. A second installer
stops with the lock path instead of racing an update. After an unhandled process
termination, confirm the original installer has stopped before removing its
.awesome-latex-skills.lock; active locks are never automatically expired.
Give your agent access to the complete skill directory, then ask:
Read awesome-latex-skills/latex-rescue/SKILL.md and follow its workflow.
In a chat-only interface, provide the entrypoint and the relevant reference files.
Pasting SKILL.md alone leaves out supporting guidance. Local editing and
compilation depend on the tools available to that agent. The legacy
agents/config.yaml files are integration hints, not a universal configuration
API for every platform.
03 / In practice
A complete manuscript workflow. Inspect the root and dependencies, build with recorded project settings, then review source changes alongside retained logs and actual PDF pages.
python -m pip install ".[pdf]"
als project check examples/full-paper/after --bundle work/project-inspection
als verify inspection work/project-inspection
als paper --output work/full-paper-run
Explore the multi-file English & Chinese case
— sections, bibliography, local style, grouped table, figure panels and appendix.
The original fails; the repaired candidate must compile. Open
work/full-paper-run/review/report.html to inspect the source diff, build evidence,
page previews and the timing protocol still awaiting the author's decision.
Project doctor & configuration · Unified change review · Formula changes
Open work/project-inspection/report.html for located issues, dependency
resolutions and next steps. Transfer this complete directory to keep its JSON
and integrity manifest together. The full runner seals both inspection reports.
| Task | Where to look |
|---|---|
| Inspect sources and shared resources | project-inspection/report.html · Nested layouts · Dependency syntax |
| Handle large files and review notes | Input sizes and retrying |
| Check class/package options and backend declarations | Loader options |
| Check database/manual bibliography keys and duplicates | Bibliography guide |
| Locate common citation commands and changed keys | Citation checks |
| Find repeated labels, missing targets and swapped range endpoints | Cross-reference checks |
| Locate formulas, incomplete pairs and operator changes | Formula review |
| Monitor selected bibliography/style inputs during builds | Build input guards |
| Check a transferred report | als verify inspection path/to/report-directory |
| Review source edits and real PDF pages | full-paper-run/review/report.html |
The review opens with file changes, literal content signals, author decisions
and build states; source diffs and complete evidence remain expandable.
Use --language zh on als review or als paper for a Chinese interface.
Inspection distinguishes comments, escaped backslashes and supported literal examples
Frequently Asked Questions
What is awesome-latex-skills?
awesome-latex-skills is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by Calix-L. LaTeX manuscript workflows with five agent skills, project diagnostics, reproducible builds and offline change review. It has 187 GitHub stars.
Is awesome-latex-skills safe to use?
Yes. awesome-latex-skills passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.
How do I install awesome-latex-skills?
Clone the repository with "git clone https://github.com/Calix-L/awesome-latex-skills" and add it to your Claude Code skills directory (see the Installation section above).
What programming language is awesome-latex-skills written in?
awesome-latex-skills is primarily written in Python. It is open-source under Calix-L on GitHub, so you can review or fork the full source.
Are there alternatives to awesome-latex-skills?
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh awesome-latex-skills against similar tools.
Comments (0)
to leave a comment.
No comments yet. Be the first to share your thoughts!
Related Skills
Reverse engineer anything with agents, from app behavior down to native binaries.
453,914
111,553
TypeScript
AI Agentsagent-skillsai
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
301,625
44,496
JavaScript
AI Agentsai-agentsanthropic
Skills for Real Engineers. Straight from my .agents directory.
250,518
21,171
Shell
AI Agents
An agentic skills framework & software development methodology that works.
234,966
20,863
Shell
AI Agentsaibrainstorming