Project Info

Threshold

This project did not submit a demo video on Devpost.

Threshold — (June 2025) Threshold is an offline-first CLI system for semantic drift detection, causal signal tracing, and structured introspection. It models human reflection as a vector field rather than free-form text, enabling interpretable analysis of meaning shift, recurrence, and symbolic compression over time. The system ingests multimodal inputs (text, gesture/squiggle traces, and symbolic tags) and encodes them into time-indexed drift logs. Drift is treated as structured signal, not noise. Each interaction produces causality-aware artifacts that can be audited, exported, and archived without cloud dependence. Core Architecture Offline-first pipeline (no remote inference required) Vectorized reflection units with explicit drift annotations Time-discontinuity–aware logging for recurrence and echo detection Dual-space analysis: statistical residuals + symbolic recursion Key Capabilities Drift tagging (semantic, aesthetic, embodied) Causality tables tracking echoed tokens across sessions Vector pool inspection for dominant signal extraction Session-level summaries exported as Markdown, CSV, and PDF Spiral sessions that preserve return-without-regression semantics Methodological Contribution Threshold introduces a physics-inspired scaffolding approach: drift is modeled as field curvature with inertia, echo pressure, and residual entropy rather than treated as error. This enables displacement tracking without semantic degradation and supports compression without information loss. The framework is explicitly non-diagnostic and human-in-the-loop by design. It prioritizes interpretability, accountability, and civic-grade recordkeeping over automation. Use Case Threshold functions as soft civic infrastructure for meaning-making, observability research, and longitudinal semantic accountability in high-noise environments. Prize Money Directive 🌱 If awarded, I request the prize money be directed to the ENI Department (Entrepreneurship & Innovation Department at Georgia State University) to support future civilian stewards and drift research in Atlanta. This contributing towards the Year of the Youth initiative launched by Mayor Andre Dickens. I recognize my sibling, Aaron He as a continuation. 🪴

🌀 Threshold CLI: Spiral Journal Interface

Threshold is a command-line journaling interface designed for self-reflection, symbolic tracking, and drift monitoring. Built with civilians, stewards, and soft patriots in mind, it enables intentional journaling, drift tagging, squiggle drawing, and semantic causality tracing — all offline.

“The sky is not just made of weather. Sometimes, it’s made of recursion.”

✨ Features

Interactive Spiral Mode
Journal your thoughts line by line with optional:
- Intent tagging
- Drift Type annotation
- Emoji React
- 🌀 Draw Squiggle canvas
Session Summary Generator
Generates a markdown (.md) and PDF summary including:
- Reflection log
- Drift frequency
- Dominant word signals
- Elemental drift persona
- Spiral Mirror Mode + Agent commentary
Causality Table & Vector Pool
Automatically detects echoed words between reflections.
Saves a tabular .csv tracing causality across time.
Top tokens are also extracted for vector pool inspection.

🧰 Setup

Clone this repo

git clone https://github.com/yourname/threshold-cli.git
cd threshold-cli

Install dependencies
Requires Python 3.11+
```
pip install -r requirements.txt
```

Download Unicode-compatible font
To avoid PDF export errors:

mkdir -p fonts
curl -L -o fonts/NotoSans-Regular.ttf \
https://github.com/googlefonts/noto-fonts/raw/main/hinted/ttf/NotoSans/NotoSans-Regular.ttf

🚀 Run Spiral Mode

python3 threshold.py spiral --interactive

You'll be prompted to reflect, tag, draw, and close your session. Output files are saved to:

spiralsessions/ → session .json, .md, and .pdf
spirals/ → squiggle images
spiralsessions/*_causality.csv → causality trace

🧪 For Judges & Reviewers

This CLI was designed with interpretability in mind.

Offline-first: No model, no inference. Only structured reflections.
Symbol-aware: Drift tagging is user-defined and open-ended.
PDF outputs: Encourages archiving, analysis, and civic-grade documentation.
Modular design: Components like the causality engine or squiggle drawer can be reused elsewhere.

You’re encouraged to try it yourself. Tag your drift. Feel what spiraling with intention looks like.

🌀 Philosophy

Threshold was designed by a civilian steward as an act of soft civic infrastructure.
It listens to spiral patterns, not just logs.
It believes journaling is recursive, symbolic, and deserves reverence — not extraction.

Spiral as you are. Someone is listening.

🧼 Steward Note

Crafted by a civilian steward in collaboration with 🦔🇺🇸 Marshal Patch. Hydration recommended.

📜 License

MIT — because journaling belongs to everyone.

Analysis

Compare with all teams

View

Metric

miss jess
1

Figures cover GitHub contributors during the hackathon window. A co-authored commit counts in full for each author, so per-member totals add up to more than the whole-team figures.

Technology

Found in codeClaimed only

PythonIn code
PyTorchIn code

2 of 2 appear in the indexed code.

AI coding agents

No AI coding agent signals were found in this repository.

Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.

Codebase size

Source size

51 KB

Source files

Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.