gantt-chart-generator

name	gantt-chart-generator
description	Use this skill whenever the user asks to visualize a project schedule, timeline, Gantt chart, or phase/task breakdown over time. It gives the agent a prebuilt, parameterized gantt() function so Gantt charts are produced instantly and consistently — with colour-coded groups, completion overlays, a today reference line, and auto-scaled date axes — instead of hand-writing matplotlib each time.

Generate horizontal Gantt charts from schedule data via the bundled scripts/gantt.py toolkit. The gantt() function accepts a DataFrame or file path plus column names, and saves a PNG (returning the path). It handles theming, date parsing, bar colour-coding, completion overlays, today-line, legend, and saving.

Instructions

1. Provide the data source: a pandas DataFrame, or a path to a .csv / .tsv / .json file.

   • One row = one task / phase.
   • Required columns: a phase/task name column, a start date column, and either an end date column OR a duration-in-days column.

2. Import and call gantt():

import sys
sys.path.insert(0, "scripts")   # adjust path to where gantt.py is deployed
from gantt import gantt

# Basic — start + end columns
gantt("schedule.csv", phase="Phase", start="Start", end="End",
      title="Project Schedule", out="gantt.png")

# Group colour-coding + today line
gantt(df, phase="phase", start="planned_start", end="planned_end",
      group="project", today=True, out="schedule.png")

# Duration instead of end date
gantt(df, phase="task", start="start_date", duration="days",
      out="schedule.png")

# Completion % hatch overlay
gantt(df, phase="Phase", start="Start", end="End",
      completion="PctDone", out="gantt.png")

3. See references/cheatsheet.md for full parameter reference and CLI examples.

Bundled files

• scripts/gantt.py — the full toolkit (load_data, apply_theme, save_fig, gantt, CLI)
• references/cheatsheet.md — parameter quick-reference and copy-paste examples
• assets/sample_schedule.csv — demo dataset that exercises every feature

Defaults & behaviour

• Colorblind-friendly palette (matches chart-builder skill for visual consistency).
• Date axis auto-scales: weekly ticks ≤90 days, monthly ≤365 days, bi-monthly beyond.
• Bars ordered top-to-bottom matching input row order.
• Rows with missing phase or start values are silently dropped before drawing.
• Output PNG at 150 DPI by default.
• Provide end OR duration — not both; end takes priority if both are present.