AutoWeave

AutoWeave is developed and maintained by Sloths Intel. It is a deterministic, browser-first CSV merge + ETL tool designed for real-world operational datasets (time logs, income records, project lists) that arrive asynchronously and inconsistently.

AutoWeave focuses on reproducibility, auditability, and low-friction workflows: validate → normalise → merge → deduplicate → summarise → visualise → export.

🌐 Website: https://autoweave.slothsintel.com

---

Contents

• Features
• Architecture
  • Tech Stack
• Local Development
  • Clone the repository
  • Run a local static server
• Data Contracts
  • Time
  • Income
  • Projects
• Merge & Validation Rules
• Quick Stats & Visualisations
• Build & Deploy
• Roadmap
• Contribution
• License
• Links

---

Features

Guided Merge (Deterministic ETL)

• Upload time, income, and project CSV files
• Validate schema before processing
• Normalise dates, ids, and numeric fields
• Deterministic join keys (no fuzzy matching)
• Deduplicate merged output
• Export a clean merged dataset

Quick Stats

• Row count
• Total time
• Total income
• Income/hour
• Project-level summaries:
  • total time
  • total income
  • income/hour

Visual Overview

• Stacked daily charts (date on x-axis)
  • Time by project
  • Income by project
  • Income/hour by project
• Floating legend (AutoTrac-style)
• Designed to be R-style readable: structured tables + consistent scales

Demo Mode (Auto-load)

• Auto-loads sample datasets from:

  • assets/technology/time_sample.csv
  • assets/technology/income_sample.csv
  • assets/technology/project_sample.csv

• User uploads override demo data.

---

Architecture

AutoWeave frontend:

├── AutoWeave
│   ├── index.html
│   ├── tech.html
│   ├── script.js
│   ├── autoweave.svg
│   └── assets
│       └── technology
│           ├── time_sample.csv
│           ├── income_sample.csv
│           └── project_sample.csv
└── README.md

Tech Stack

Frontend

• HTML + CSS
• Vanilla JavaScript
• CSV parsing (client-side)
• Browser rendering for tables + charts

Backend

• Backend: FastAPI (validation + persistence)
• Database: PostgreSQL (versioned datasets + lineage)

---

Data Contracts

AutoWeave expects three datasets: time, income, and projects (optional).

Time

Required columns:

• project_id (string)
• work_date (date; ISO preferred)

Recommended columns (used to compute/validate duration):

• start_time (time)
• end_time (time)
• duration (number; hours or minutes depending on your export — AutoWeave normalises to hours)

Income

Required columns:

• project_id (string)
• work_date (date)
• income (number)

Projects

Required columns:

• project_id (string)
• project (string; display name)

---

Merge & Validation Rules

Normalisation

• Trim whitespace on all identifiers
• Coerce numeric fields (income, duration)
• Normalise work_date to a consistent date key
• Treat missing values explicitly (no silent inference)

Invalid Time Rows

• Drop time entries where both start_time and end_time are missing/unparseable and duration is missing/unparseable.

Join Strategy

Deterministic join key:

project_id + work_date

Income aligns to time entries by exact match on this key.

Deduplication

Merged output is deduplicated using a deterministic composite key (implementation-defined), typically:

project_id + work_date + duration + income

AutoWeave does not perform fuzzy matching.

---

Quick Stats & Visualisations

Quick Stats

Computed from the merged dataset:

• Total rows
• Total time
• Total income
• Income per hour
• By-project summaries:
  • total time
  • total income
  • income per hour

Visualisations

Stacked daily charts:

• Time by project
• Income by project
• Income/hour by project

All charts:

• Use a consistent project color mapping
• Include a floating legend
• Default to most recent dates available in the merged dataset

---

Roadmap

• export audit metadata
• predictive hooks (AutoPred)

---

Contribution

Maintained by Sloths Intel GitHub, and Daddy Sloth GitHub.

---

License

© 2026 Sloths Intel.

A trading name of Sloths Intel Ltd Registered in England and Wales (Company No. 16907507).

MIT License.

---

Links

• AutoWeave Website
• AutoWeave GitHub
• Company homepage

⬆️