Skip to content

grainulation/harvest

Harvest

npm version npm downloads license node CI Explore on DeepWiki

Are your decisions getting better?

Harvest is the analytics layer for research sprints. It looks across sprints to find patterns, score predictions, and surface knowledge that's gone stale.

Install

npm install -g @grainulation/harvest

Or use directly:

npx @grainulation/harvest analyze ./sprints/

What it does

  • Cross-sprint analysis -- claim type distributions, evidence quality, recurring themes
  • Prediction calibration -- score past estimates against actual outcomes
  • Decision patterns -- what research approaches lead to better results?
  • Knowledge decay -- which old claims need refreshing before they mislead you?
  • Sprint velocity -- how long do sprints take, where do they stall?
  • Retrospective reports -- dark-themed HTML reports for the team

Quick start

# Cross-sprint claim analysis
harvest analyze ./sprints/

# Score predictions against outcomes
harvest calibrate ./sprints/

# Detect decision patterns and anti-patterns
harvest patterns ./sprints/

# Find stale claims that need refreshing
harvest decay ./sprints/ --days 60

# Sprint timing and phase analysis
harvest velocity ./sprints/

# Generate a full retrospective HTML report
harvest report ./sprints/ -o retrospective.html

# All analyses in one pass
harvest trends ./sprints/ --json

# Start the live dashboard (SSE updates, dark theme)
harvest serve --root ./sprints/ --port 9096

# Connect to farmer for mobile monitoring (farmer's default port is 9090)
harvest connect farmer --url http://localhost:9090

Data format

Harvest reads standard wheat sprint data:

  • claims.json -- array of typed claims with id, type, evidence, status, text, created
  • compilation.json -- compiled sprint state (optional, enriches analysis)
  • Git history on claims.json -- used for velocity and timing analysis

Point harvest at a directory containing sprint subdirectories, or at a single sprint directory.

Design

  • Reads, never writes -- harvest is a pure analysis tool; it won't modify your sprint data
  • Git-aware -- uses git log timestamps for velocity analysis when available
  • Composable -- each module (analyzer, calibration, patterns, decay, velocity) works independently

Zero third-party dependencies

Depends only on @grainulation/barn (internal ecosystem utilities); no third-party runtime deps. Node built-ins otherwise.

Part of the grainulation ecosystem

Tool Role
wheat Research engine -- grow structured evidence
farmer Permission dashboard -- approve AI actions in real time
barn Shared tools -- templates, validators, sprint detection
mill Format conversion -- export to PDF, CSV, slides, 26 formats
silo Knowledge storage -- reusable claim libraries and packs
harvest Analytics -- cross-sprint patterns and prediction scoring
orchard Orchestration -- multi-sprint coordination and dependencies
grainulation Unified CLI -- single entry point to the ecosystem

Releases

See CHANGELOG.md for release history.

License

MIT

About

Sprint analytics and claim visualization. Charts, timelines, and evidence reports.

Topics

Resources

License

Code of conduct

Contributing

Security policy

Stars

Watchers

Forks

Contributors