table-generation — Claude Code skill: LaTeX tables from results
Turns JSON or CSV experiment results into booktabs-styled LaTeX tables with bold best values, captions and labels.
npx skills add lingzhi227/agent-research-skills --skill table-generation -a claude-codeWhat table-generation does
The skill reads a results file and builds a paper-ready table from it. The table type comes in as an argument: comparison for main results, ablation for model variants, descriptive for dataset statistics, multi-dataset for a methods by datasets by metrics layout. The best value in each column gets bold, the runner-up can be underlined with --underline-second, and p-value stars come from --significance.
Under the hood there is a results_to_table.py script you call from the shell with the data path, table type, bold-best mode and caption. A references/table-templates.md file holds templates and examples.
The generated code expects booktabs, multirow, multicol and threeparttable in your preamble — without them it will not compile. The script pulls numbers only from the source you hand it, so a mismatch with your experiment logs shows up right away. Worth checking the data before generation rather than after.
The skill sits in a research pipeline: data analysis and experiment code feed into it, paper writing and compilation come after.
What it helps with
- build a main comparison table across methods
- lay out an ablation table with component checkmarks
- print dataset statistics with units
- fit methods, datasets and metrics into one wide table
- add p-value stars and underline the second-best result
Example prompts for the agent
- Generate a comparison table from results.json and bold the best result
- Build an ablation table from variants.csv with checkmarks for components
- Create a descriptive dataset table with a caption and label
How to install table-generation
1Skills CLI (Vercel)
npx skills add lingzhi227/agent-research-skills --skill table-generation -a claude-code2Manually: put the files in place
~/.claude/skills/table-generation/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/table-generation/
1Skills CLI (Vercel)
npx skills add lingzhi227/agent-research-skills --skill table-generation -a codex2Manually: put the files in place
.agents/skills/table-generation/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add lingzhi227/agent-research-skills --skill table-generation -a cursor2Manually: put the files in place
.cursor/skills/table-generation/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install lingzhi227/agent-research-skills table-generation2Manually: put the files in place
.github/skills/table-generation/SKILL.md1Skills CLI (Vercel)
npx skills add lingzhi227/agent-research-skills --skill table-generation -a gemini-cli2Platform CLI
gemini skills install https://github.com/lingzhi227/agent-research-skills --path skills/table-generationManually: put the files in place
.devin/skills/table-generation/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add lingzhi227/agent-research-skills --skill table-generation -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about table-generation
How do I install table-generation in Claude?
Run npx skills add lingzhi227/agent-research-skills --skill table-generation -a claude-code. The skill file then lands at ~/.claude/skills/table-generation/SKILL.md.
What does table-generation do?
It converts JSON or CSV experiment results into LaTeX tables with booktabs rules, bold best values, a caption and a label.
Which LaTeX packages do the tables need?
booktabs, multirow, multicol and threeparttable. The last one is for table notes.
