How do I install the Annotating CSV skill?

Run npx skillmds@latest add diegosouzapw/annotating-csv in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Annotating CSV safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Annotating CSV?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Annotating CSV free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Annotating CSV?

diegosouzapw (@diegosouzapw) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Annotating CSV by diegosouzapw — Agent Skill

Name: Annotating CSV
Rating: 4.2 (5 reviews)
Author: diegosouzapw

Annotating CSV

Annotates CSV rows using OpenAI by applying a prompt to each row and adding a new column with AI-generated results. Use when the user wants to classify, extract, summarize, or add AI-generated annotations to CSV data. Triggers on mentions of CSV annotation, row-by-row AI processing, bulk classification, or adding AI columns to spreadsheet data.

diegosouzapw Updated Sep 8, 2026 54 repo stars

File contents

CSV Annotation with OpenAI

Adds AI-generated annotations to CSV files by processing each row with a prompt.

Quick Start

uv run annotate-csv.py "Classify sentiment as positive/negative/neutral" reviews.csv

Output: reviews_annotated.csv with new annotation column.

Dependencies install automatically on first run.

Core Usage

uv run annotate-csv.py <prompt> <csv_file> [options]

Option	Description	Default
`-o, --output`	Output file path	`<input>_annotated.csv`
`-c, --column`	Annotation column name	`annotation`
`--context`	Columns to use (`all` or comma-separated)	`all`
`--model`	OpenAI model	`gpt-5-mini`
`--parallelism`	Parallel workers	`10`
`--id-column`	Column for progress display	-

Prompt Sources

Prompts can be inline or from a file:

# Inline prompt
python annotate-csv.py "Extract the main topic" data.csv

# From file (for complex prompts)
python annotate-csv.py prompt.txt data.csv

Common Patterns

Classification:

python annotate-csv.py "Classify sentiment: positive, negative, or neutral" feedback.csv -c sentiment

Extraction:

python annotate-csv.py "Extract the product name mentioned" reviews.csv -c product --context "review_text"

Summarization:

python annotate-csv.py "Summarize in one sentence" articles.csv -c summary

See EXAMPLES.md for more patterns and prompt templates.

Environment Setup

Requires OPENAI_API_KEY in .env file:

OPENAI_API_KEY=sk-...

Run with uv (recommended):

uv run annotate-csv.py "Your prompt" data.csv

Or install dependencies first:

uv pip install -r requirements.txt
python annotate-csv.py "Your prompt" data.csv

Workflow

Preview data: Check CSV structure and columns
Choose context: Decide which columns the AI needs to see
Write prompt: Be specific about expected output format
Test on subset: Try on a few rows first if large dataset
Run full annotation: Process the complete file
Validate results: Spot-check annotation quality

Writing Effective Prompts

Be explicit about output format (single word, phrase, sentence)
List valid categories for classification tasks
Specify what to do with ambiguous cases
Keep prompts focused on one task

See PROMPTS.md for prompt-writing guidance.

Troubleshooting

Issue	Solution
API key error	Check `.env` file has `OPENAI_API_KEY`
Column not found	Verify column names match CSV exactly
Rate limits	Reduce `--parallelism` to 5 or lower
Empty annotations	Check context columns have data

---name: annotating-csvdescription: Annotates CSV rows using OpenAI by applying a prompt to each row and adding a new column with AI-generated results. Use when the user wants to classify, extract, summarize, or add AI-generated annotations to CSV data. Triggers on mentions of CSV annotation, row-by-row AI processing, bulk classification, or adding AI columns to spreadsheet data.---# CSV Annotation with OpenAIAdds AI-generated annotations to CSV files by processing each row with a prompt.## Quick Start```bashuv run annotate-csv.py "Classify sentiment as positive/negative/neutral" reviews.csv```Output: `reviews_annotated.csv` with new `annotation` column.Dependencies install automatically on first run.## Core Usage```bashuv run annotate-csv.py <prompt> <csv_file> [options]```| Option | Description | Default ||--------|-------------|---------|| `-o, --output` | Output file path | `<input>_annotated.csv` || `-c, --column` | Annotation column name | `annotation` || `--context` | Columns to use (`all` or comma-separated) | `all` || `--model` | OpenAI model | `gpt-5-mini` || `--parallelism` | Parallel workers | `10` || `--id-column` | Column for progress display | - |## Prompt SourcesPrompts can be inline or from a file:```bash# Inline promptpython annotate-csv.py "Extract the main topic" data.csv# From file (for complex prompts)python annotate-csv.py prompt.txt data.csv```## Common Patterns**Classification:**```bashpython annotate-csv.py "Classify sentiment: positive, negative, or neutral" feedback.csv -c sentiment```**Extraction:**```bashpython annotate-csv.py "Extract the product name mentioned" reviews.csv -c product --context "review_text"```**Summarization:**```bashpython annotate-csv.py "Summarize in one sentence" articles.csv -c summary```**See [EXAMPLES.md](EXAMPLES.md) for more patterns and prompt templates.**## Environment SetupRequires `OPENAI_API_KEY` in `.env` file:```OPENAI_API_KEY=sk-...```Run with uv (recommended):```bashuv run annotate-csv.py "Your prompt" data.csv```Or install dependencies first:```bashuv pip install -r requirements.txtpython annotate-csv.py "Your prompt" data.csv```## Workflow1. **Preview data**: Check CSV structure and columns2. **Choose context**: Decide which columns the AI needs to see3. **Write prompt**: Be specific about expected output format4. **Test on subset**: Try on a few rows first if large dataset5. **Run full annotation**: Process the complete file6. **Validate results**: Spot-check annotation quality## Writing Effective Prompts- Be explicit about output format (single word, phrase, sentence)- List valid categories for classification tasks- Specify what to do with ambiguous cases- Keep prompts focused on one task**See [PROMPTS.md](PROMPTS.md) for prompt-writing guidance.**## Troubleshooting| Issue | Solution ||-------|----------|| API key error | Check `.env` file has `OPENAI_API_KEY` || Column not found | Verify column names match CSV exactly || Rate limits | Reduce `--parallelism` to 5 or lower || Empty annotations | Check context columns have data |

Frequently asked questions