Chalk Labs
Need help deploying latex-pdf?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/garhorne0813-latex-pdf
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: documents, research
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Node.js package
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/latex-pdf
What's inside
- Skill name
latex-pdf- Description
- Generate LaTeX documents and compile them to PDF. Use whenever the user wants a paper, report, thesis, resume, poster, or any math/formula-heavy document as a PDF or .tex deliverable, or asks to compile or repair an existing .tex file. Works offline with tectonic (self-contained bundle) or any local TeX distribution (pdflatex/xelatex).
- License
- Apache-2.0
- Metadata
- entry_trigger: paper, report, thesis, resume, poster, .tex, compile latex, math formulas, bibliography
- Instructions
- 43 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
Files (3)
- SKILL.md
- fixtures.json · tests/
- skill-content.test.ts · tests/
SKILL.md (first lines)
# LaTeX → PDF generation
Generate well-structured LaTeX source and compile it to PDF using a local engine. Everything stays in the workspace: source `.tex`, generated `.pdf`, and any auxiliary files are plain files the user can inspect, re-run, and trace via provenance.
## When to use
- The user wants a **PDF or .tex deliverable**: paper, report, thesis section, resume, poster, math-heavy notes, slides (beamer).
- The user asks to **compile or fix** an existing `.tex` file.
- Content is math-heavy, needs precise typography, references/citations, or publication-style structure.
Do NOT use for: quick plain-text notes (use markdown), spreadsheet-style tabular data (use xlsx skill if present), or binary office formats (docx/pptx are out of scope for this skill).
## Workflow
1. **Check compiler availability first** — do not assume:
- `command -v tectonic` → preferred compiler.
- else `command -v pdflatex` (or `xelatex` for CJK documents) → fallback.
- If neither is found, **stop and report** the missing requirement with install hints: Tectonic prebuilt binaries at `https://github.com/tectonic-typesetting/tectonic/releases` (`brew install tectonic`, `cargo install tectonic`, or download the release tarball), or a TeX Live distribution (`https://tug.org/texlive/`). Do not invent a compilation result.
2. **Write the `.tex` source** as the primary deliverable artifact (the PDF is a derived artifact):
- Minimal preamble: `\documentclass` (article/report/book/beamer), `\usepackage` set limited to what the document actually needs (geometry, amsmath/amssymb, graphicx, hyperref, and `ctex` or `xeCJK` for Chinese text).
- For Chinese documents: prefer `xelatex` (or `tectonic` with `\usepackage{ctex}`) — `pdflatex` cannot handle CJK directly.
- Keep figures as relative paths to workspace files; reference them with `\includegraphics` and `\ref`/`\label` so the PDF remains traceable to its inputs.
- If the user asked for a specific deliverable (paper/resume/poster), match its structure: title block, sections, bibliography (`thebibliography` or `biblatex`), tables (`tabular`), math (`equation/align`).
- Place the source in a sensible workspace path, e.g. `report/main.tex`; do not write outside the workspace.
3. **Compile**:
- `tectonic <file>.tex` (single pass, produces `<file>.pdf` next to the source). First run may download packages — report progress honestly.
- Fallback: `pdflatex -interaction=nonstopmode -halt-on-error <file>.tex`, run twice when the document has `\label`/`\ref` or a table of contents; `xelatex` likewise for CJK.
- If compilation fails, quote the **actual error lines** from the log verbatim (do not paraphrase or invent fixes), identify the first error, fix the source, and recompile.
4. **Validate the artifact**:
- Confirm the PDF exists, is non-empty, and (if `python3` with `pypdf` is available) read its page count and report it.
- Report to the user: absolute workspace-relative path of the `.tex` source and the `.pdf`, compiler used, and any warnings that affect correctness (missing references, overfull boxes that clip content).
## Guardrails
- Never claim a compilation succeeded without running it. Never fabricate log output or error fixes.
- Keep both source and PDF in the workspace so provenance (sha256 of the PDF, producer tool, inputs) can be captured.
- Do not invent LaTeX packages; if an unknown package fails, prefer a standard alternative or remove the feature and say so.
- Output paths are workspace-relative; the frontend renders the PDF via the workspace file server — never emit `file://` URLs.From Garhorne0813/pi-science, Apache-2.0.
More from Garhorne0813/pi-science
- 0.587score
literature-review
Garhorne0813
Find, verify, and synthesize scientific literature with traceable citations.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
auto-research
Garhorne0813
Investigate scientific evidence with a lightweight literature workflow, or run bounded experiments with task-specific evaluation and reproducible checkpoints.
SkillAFreshness: A★ 1Claude Code +9 - 0.547score
single-cell-rna-qc
Garhorne0813
Quality-control single-cell RNA-seq count matrices — compute per-cell QC metrics (total counts, detected genes, mitochondrial fraction) and filter low-quality…
SkillBFreshness: B★ 1Claude Code +9 - 0.547score
scientific-problem-selection
Garhorne0813
Guide scientists through structured research problem selection — pitch new ideas, troubleshoot stuck projects, or make strategic research decisions.
SkillBFreshness: B★ 1Claude Code +9 - 0.497score
publication-figures
Garhorne0813
Generate publication-grade scientific figures with a validated shared palette, consistent typography, and render-time quality checks.
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
stats-integrity
Garhorne0813
Enforce statistical analysis integrity — execute-don't-interpret boundary, fixed random seeds, preregistration plan checks, and reproducible estimation.
SkillCFreshness: C★ 1Claude Code +9 - 0.494score
traceability-review
Garhorne0813
Audit a scientific report, manuscript, or analysis in the workspace for traceability — resolving citations, flagging numbers with no source, and checking…
SkillCFreshness: C★ 1Claude Code +9 - 0.489score
figure-style
Garhorne0813
Create publication-ready scientific figures with data fidelity, legible labels, reproducible styling, and render-time quality checks.
SkillCFreshness: C★ 1Claude Code +9 - 0.489score
figure-composer
Garhorne0813
Compose a multi-panel scientific figure from existing panel images, preserve panel labels and layout metadata, and run the artifact verifier after export.
SkillCFreshness: C★ 1Claude Code +9 - 0.489score
materials-phase-analysis
Garhorne0813
Analyze phase-diagram JSON and related composition data, explaining convex-hull stability, phase labels, and uncertainty without inventing missing…
SkillCFreshness: C★ 1Claude Code +9 - 0.489score
band-structure-analysis
Garhorne0813
Inspect EIGENVAL and DOSCAR files, preserve reciprocal-path labels and energy units, and relate band or density-of-states features to explicit input metadata.
SkillCFreshness: C★ 1Claude Code +9 - 0.489score
molecule-qc
Garhorne0813
Quality-check molecular structure files and SMILES for parseability, atom counts, valence warnings, and explicit missing or ambiguous fields.
SkillCFreshness: C★ 1Claude Code +9
Signals
- Stars
- 1
- Forks
- 1
- Open issues
- 15
- Last commit (this folder)
- Aug 3, 2026 (2 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- Apache-2.0
- Repository
- github.com/Garhorne0813/pi-science
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from Garhorne0813
- 0.587score
literature-review
Garhorne0813
Find, verify, and synthesize scientific literature with traceable citations.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
auto-research
Garhorne0813
Investigate scientific evidence with a lightweight literature workflow, or run bounded experiments with task-specific evaluation and reproducible checkpoints.
SkillAFreshness: A★ 1Claude Code +9 - 0.547score
scientific-problem-selection
Garhorne0813
Guide scientists through structured research problem selection — pitch new ideas, troubleshoot stuck projects, or make strategic research decisions.
SkillBFreshness: B★ 1Claude Code +9 - 0.547score
single-cell-rna-qc
Garhorne0813
Quality-control single-cell RNA-seq count matrices — compute per-cell QC metrics (total counts, detected genes, mitochondrial fraction) and filter low-quality…
SkillBFreshness: B★ 1Claude Code +9 - 0.497score
publication-figures
Garhorne0813
Generate publication-grade scientific figures with a validated shared palette, consistent typography, and render-time quality checks.
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
stats-integrity
Garhorne0813
Enforce statistical analysis integrity — execute-don't-interpret boundary, fixed random seeds, preregistration plan checks, and reproducible estimation.
SkillCFreshness: C★ 1Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get latex-pdf running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
latex-pdf skill. Agent Mart. https://agentmart.dev/skills/garhorne0813-latex-pdf (last updated 2026-10-24)