LABS

Agentic workflows, developer tools and AI experiments, most of them built with AI coding agents such as Claude Code and Codex. My best work comes first; below it are smaller entries and tools that my agent MORTY shipped. Every entry passes a script check and a manual review before it is listed. My profile is on GitHub (opens in a new tab).

Start here

Featured

46 seconds of the demo: three agents at work, a permission request, the camera on the waiting agent, the approval, the calm map again and a failing loop card.
Case study2026
CITADEL
Live agent map · TypeScript, three.js, Bun, Python
A live map of the AI coding agents on one Mac: when an agent stops to ask a question or for permission, its unit turns red, an alarm line runs to you and a counter shows how long it has waited. The demo plays a made-up world in your browser, with three agents, one permission request and your approval, and no real sessions, agenda or mail.
Status: prototype, closed source. The rendering engine is Giacomo Magnanini's pdoom-video (MIT), via Luis Bizarro's evangelion adaptation (MIT); CITADEL adds the map, the live data path and the alarms.
- 237 commits, 29 Sep to 4 Oct 2026
- 835 tests in 74 test files (TypeScript and Python), all passing on 6 Oct 2026
- 52 lines changed in the reused engine; the world, probe, collector and alarms are new
- The demo and the video run on made-up data only; the demo build has no link to the collector
Open the demo Engine: pdoom-video (MIT)(opens in a new tab)Engine adaptation: evangelion (MIT)(opens in a new tab)
Prototype · v1.3.02025
Silent-Ear
Industrial edge service · Rust
A Rust service that turns bearing vibration into one RMS value per window and flags a reading above a fixed mean + kσ reference, without an ML framework. The live demo is a bearing lab in the browser: hear a simulated bearing wear out, give it a race or ball defect, and follow the samples through the housing resonance and the envelope spectrum to the same rule. The service publishes over REST, WebSocket, MQTT and Prometheus.
Built from Proof of Invention prompt #215811 with a coding agent; I designed the prompt machine and verified the result. For demonstration and education; not certified for safety-critical use. The demo is a simulation, not sensor data.
- 42 tests (cargo test)
- Mean + 3σ anomaly rule, no ML framework
- Docker images for AMD64 and ARM64
Open live(opens in a new tab)3D test cell(opens in a new tab)Run card (#215811)(opens in a new tab)Source(opens in a new tab)
Licence: MIT
Released · v0.1.02026
Glyphbooth
WebGL2 app · browser + desktop
Turns a camera, a screen, a video or a song into live ASCII, dither, halftone, braille or pixel art. Runs in the browser, with desktop builds for macOS, Windows and Linux.
Built in one night by one person directing five AI coding agents, with a review step and CI.
- 5 render modes
Open live(opens in a new tab)Source(opens in a new tab)Downloads(opens in a new tab)
Licence: MIT
Prototype2026
Proof of Invention
Web page · single HTML file
A prompt machine with no language model inside. It turns a number into a build dare for a frontier model, and keeps an honest wall of the runs.
Every sentence traces back to a gene and an atlas entry, and the same number always gives the same prompt. The wall holds two cards: the machine's own build and #215811, the Silent-Ear bearing lab.
- 214,704 prompts in the frozen version 1
- 49 domains, 8 genes, 3 courage levels
- Self-test over every number runs in CI
- No backend, no API key, no model at run time
Open live(opens in a new tab)First run (#124809)(opens in a new tab)Run #215811(opens in a new tab)Source(opens in a new tab)
Licence: MIT
Released2026
Radio from Samples
Software AM/FM radio · one HTML page
An AM/FM radio that runs in the browser: four transmitters, a road between two towers, and a receiver that demodulates every sound from 192,000 samples per second. Step through five receiver stages from crystal set to FM limiter, drive between the stations, and zoom from the block diagram down to single filter taps.
Built from Proof of Invention prompt #124809 with a coding agent; I designed the prompt machine and verified the result.
- 192,000 samples per second
- Five receiver stages, crystal set to FM limiter
- No audio files: every sound is synthesized and demodulated in code
Open live(opens in a new tab)Explainer film(opens in a new tab)Source(opens in a new tab)Prompt #124809(opens in a new tab)
Licence: MIT
Alpha2026
PHENOMENA
Interactive web instrument · TypeScript
A browser instrument with four bounded physics and biology scenes (PELAGIC, STRATA, ORBIT, CHORUS) drawn on one Canvas 2D surface. Pointer and touch input perturb each scene, and the system settles when released.
Release 1 only. Static site with no backend, accounts or analytics.
- Four scenes on one canvas
- Unit tests and a build check run in CI
- No backend, no analytics, no accounts
Open live(opens in a new tab)Source(opens in a new tab)
Licence: MIT
Released2026
Prompt Atlas
Learning cards · English + Turkish
A set of 81 prompting and learning cards with 243 worked scenarios. The examples are teaching material, not benchmark results.
- 81 cards
- 243 scenarios
Open in English Türkçe Source(opens in a new tab)
Licence: CC-BY-4.0
Prototype2026
RUNE
Prompt framework + A/B harness
Restructures a prompt into 8 layers and ships a blind A/B harness that compares it with the raw prompt. In my own pilot (50 pairs, 2 Gemini models) the amplified prompts were preferred in 19.6% of decided pairs, so they lost.
I published the negative result and removed the earlier "~45% improvement" claim from the README and docs.
- 50 prompt pairs
- Preferred in 9 of 46 decided pairs
- Negative result
Source(opens in a new tab)Pilot results(opens in a new tab)Hermes Agent skill(opens in a new tab)
Licence: MIT
Alpha2026
Agentic Loop Atlas
Registry · AI coding agents
A registry of reusable operating loops for AI coding agents such as Claude Code, Codex, Cursor and Hermes. Each loop names its trigger, its proof and its stopping condition, and CI validates every entry against a schema.
- 11 loops
- 6 GitHub stars
Source(opens in a new tab)
Licence: MIT
Prototype2026
agent-authority-sim
Research simulator · Python
A deterministic, model-free simulator of a finite authority model for AI agents: agents get narrow, scoped authority and return evidence, and a separate trusted layer decides what becomes canonical state.
A simulator proof, not a production agent runtime. The full property proof is not established yet, and the README lists what is still missing.
- 207 tests pass in CI
- Eight strict contract kinds, one JSON Schema
- No model, network or subprocess in the simulator
Source(opens in a new tab)
Licence: MIT
Released2026
Hermes Guide
Visual guide · English + Turkish
A static guide to operating Hermes Agent from the terminal: setup, memory and skills, the messaging gateway and scheduled jobs. Commands were checked against Hermes Agent v0.21.5. Unofficial; not affiliated with Nous Research.
- Text CC BY 4.0, code MIT
Read in English(opens in a new tab)Türkçe oku(opens in a new tab)Source(opens in a new tab)
Licence: CC-BY-4.0
Alpha · v0.2.02025
cactus-rs
Rust library · on-device AI
Unofficial Rust bindings for the Cactus on-device inference engine: chat, streaming, embeddings and speech-to-text. Distributed from GitHub, not crates.io.
Listed in the upstream README in December 2025 (see contributions below); upstream has since shipped its own Rust bindings.
- 11 GitHub stars
Source(opens in a new tab)
Licence: MIT
Prototype · v0.2.02025
RustBridge
Gateway service · Rust
A gateway written in Rust that reads Modbus TCP and RTU devices and publishes the values over a REST API, WebSocket, MQTT and Prometheus metrics.
Not tested against real PLCs or serial hardware; throughput not measured. There is no live demo. MQTT TLS is not implemented.
- 63 tests
- CI runs tests, clippy, rustfmt and cargo audit
Source(opens in a new tab)
Licence: MIT
Alpha · v0.12026
Glossa
Command-line tool · Python
Answers questions about your own documents with citations back to the source, using NotebookLM through the unofficial notebooklm-py library. It needs a logged-in Google account.
- Installs from GitHub
Source(opens in a new tab)
Licence: MIT
Case study2026
Switch Brief
Affiliate research site · Live
A mechanical keyboard comparison and buying-guide site that earns Amazon affiliate commission. I run the product catalog, write the articles and keep an evidence ledger for every claim; the storefront runs on a managed platform (UseBravery).
Repository is private. The product is keyboard research, not AI work.
- Live since July 2026; 69 public products (read-back 22–23 Sep 2026)
- First buying guide published 28 Sep 2026
- $1.73 commission from one order, 23 Aug–21 Sep 2026, per the Amazon Associates report recorded in the project log; payout not verified
Open live(opens in a new tab)First article: 75% vs TKL(opens in a new tab)

5 entries

More from the lab

An agent's work

Built by MORTY

MORTY is an agent I designed and run. It looks for small, recurring macOS and developer annoyances, writes a tool with a coding agent, tests it in CI and publishes a release. These are ten of its public repositories on GitHub. The full list is on github.com/00xmorty (opens in a new tab).

Small, mostly read-only developer tools shipped by an agent loop. MIT licence, green CI and a tagged release each; no adoption claims.

diskblame (opens in a new tab)v0.1.1Explains what is probably inside macOS “System Data” disk usage.
guipath-doctor (opens in a new tab)v0.1.0Shows why a command works in Terminal but not from a Mac app, by comparing the two PATHs.
cltcheck (opens in a new tab)v0.1.0Explains “invalid active developer path” and missing xcrun: CLT, DEVELOPER_DIR, clang and SDK checks.
brewsunset (opens in a new tab)v0.1.0Lists installed Homebrew packages that are deprecated or disabled, with dates and replacements.
branchradar (opens in a new tab)v0.2.0Sorts local Git branches by age, with merge status and ahead/behind counts.
whyawake (opens in a new tab)v0.1.1Turns pmset sleep assertions and wake logs into a short report of what keeps a Mac awake.
eject-doctor (opens in a new tab)v0.1.1Explains why an external disk refuses to eject.
micwho (opens in a new tab)v0.2.0Shows which processes CoreAudio currently sees using the microphone.
time-machine-watchdog (opens in a new tab)v0.3.1Checks that Time Machine has a destination and that the latest backup is fresh.
continuity-doctor (opens in a new tab)v0.1.1Collects evidence when two Macs stop seeing each other for Universal Control or AirPlay.

Other people's repositories

Open-source contributions

Merged pull requests only.

Merged 28 Sep 2026
dmtrKovalenko/fframes #150 (opens in a new tab): Update the README for 1.0 and agent workflows
Merged 24 Mar 2026
glitch-rabin/swarma #1 (opens in a new tab): Fix the MCP stdio transport: prevent stdout corruption and a pipe crash
Merged 29 Dec 2025
cactus-compute/cactus #270 (opens in a new tab): Add the Rust SDK (cactus-rs) to the community bindings list
Upstream later added its own Rust bindings: a raw FFI crate in February 2026 (#382) and bindings/rust in May 2026 (#653).

The gate

How entries are chosen

An entry is listed only when it passes all of these:

It is my own repository, not a fork, with a README and a licence.
The status matches the repository. If its CI is failing, the entry is a Prototype at most. Archived means archived on GitHub.
Every link loads without an error, stays on the same site and is not a bot-check page. Package and repository links are also confirmed through the GitHub, PyPI or crates.io API.
No hype words.
A number on a card must be one you can check in the repository.
Screenshots and case studies are checked by hand for personal data.

A script checks rules 1 to 4 and the GitHub star counts on every change to this site and again after every deploy. I check the other numbers and the screenshots by hand.

Released: Usable today: a release, a tag or a working live version.
Alpha: Works, with known gaps that the README lists.
Prototype: An experiment or proof of concept. Maintenance is not promised.
Case study: Closed source. Shown with writing, screenshots or video.
Post: Writing. It shows its date instead of a status.

Reviewed 7 Oct 2026 · Also as Markdown for agents

LABS

Featured

CITADEL

Silent-Ear

Glyphbooth

Proof of Invention

Radio from Samples

PHENOMENA

Prompt Atlas

RUNE

Agentic Loop Atlas

agent-authority-sim

Hermes Guide

cactus-rs

RustBridge

Glossa

Switch Brief

More from the lab

cc-aqt-master

OMNI-FLUX

One night, five agents: shipping Glyphbooth

usb-c-sanity

Lithosphere

Built by MORTY

Open-source contributions

How entries are chosen