🏙️ ProtocolCity-BluePrint
Pre-release (0.1.x). The roof of the ProtocolCity suite — WorkLane (desk) · WorkForce (roster) · this BluePrint (map + suite CLI). Expect sharp edges; file issues.
The blueprint for running your own city of AI agents — and the installable suite that makes Map · Desk · Roster real on a laptop.
Install the suite (recommended)
brew install protocolcity/tap/protocolcity
# Path is *your* workspace name (Developer, my-city, work, … — no forced default).
protocolcity found ~/my-city
# Optional: --project <name> for one new folder at found; or drop existing
# folders under the root and `protocolcity adopt ~/my-city <folder>`.
protocolcity serve --with-engines
# → http://127.0.0.1:8801/ (Explorer · Desk · Agents)
That one formula installs the BluePrint CLI and pulls WorkLane + WorkForce from PyPI. Product source repos stay separate; install does not require cloning them.
Or: pip install protocolcity protocolcity-worklane protocolcity-workforce
Spec + paper
Protocol City is also a specification for organizing a workspace so humans and AI agents from any vendor can build together, in parallel, accountably. You can adopt the paper path with a text editor and the templates below.
Every bot is a worker, every
.mdis a law, and every contributor becomes a citizen.
Why
Agents are brilliant and unaccountable. Every vendor wants orchestration to live inside its own runtime, in its own config format. Decisions evaporate in chat windows. The Blueprint is the opposite bet: coordination as a protocol — files, tickets, and signed work — durable, auditable, human-readable, and owned by you.
WorkLane gives you the desk. WorkForce gives you the workforce. The Blueprint gives you the map.
What's inside
| Document | What it is |
|---|---|
| CHARTER.md | The spec: definitions, the three layers, the four levels of law, the compliance test, and the founding path |
| MANIFESTO.md | Why the city exists — what we believe and the wager we're making |
| FOUNDING.md | Day one: find your neighborhoods, copy the templates, pass the compliance test — an afternoon, start to finish |
| RUNNING.md | Every day after: the loops, the five citizen duties, the flywheel — and what runs itself |
| templates/ | Fill-in-the-blank law files for all four levels, plus vendor pointers |
| example/ | Riverside — a complete minimal city (one root, one neighborhood, one worker), every template filled in |
The short version
A city is your whole operation. Each project is a neighborhood. Work moves only by ticket through the work-order desk. Your agents are the workforce — any vendor, each with a registered identity that signs everything it does. The rules are laws: plain markdown files at four levels, from city-wide law down to a single worker's shift instructions.
A project is protocol-compliant when three things are true:
- Its law is written — an
AGENTS.mdat its root. - Its work is ticketed — filed at the desk, explicit scope.
- Its workers sign — every action carries a registered identity.
Start with one neighborhood, one worker, one law — FOUNDING.md walks you through it, and there's nothing to install: adopting the Blueprint is copying templates and filling them in.
Status
v0.1.x pre-release. Install path (Homebrew / PyPI) is live. Suite UX is still sharpening. The Blueprint is extracted from a working city: one operator, one machine, multiple products, agents from several vendors. Expect the spec and the shell to move quickly.
License
CC BY 4.0 — use it, adapt it, build your city on it, with attribution.
1 maintainer
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file protocolcity-0.1.7.tar.gz.
File metadata
- Download URL: protocolcity-0.1.7.tar.gz
- Upload date:
- Size: 357.5 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via:
twine/6.2.0 CPython/3.9.6
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
126548ac0f75dde6387ee19edaf8bf84af301089a6693075b59c71878ac25984
|
|
| MD5 |
2ea8d3df274951efa37a1b314deb8064
|
|
| BLAKE2b-256 |
f15ce4bfeda8937765d46ed9878a5a626fbeca69841e62a7f641893ccbe27eed
|
File details
Details for the file protocolcity-0.1.7-py3-none-any.whl.
File metadata
- Download URL: protocolcity-0.1.7-py3-none-any.whl
- Upload date:
- Size: 379.8 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via:
twine/6.2.0 CPython/3.9.6
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
c3605fec0424c9b6292c28bab2806c19e3ebfa551b94ed0894d794de029dc572
|
|
| MD5 |
2a35c0fc04dd9927792e0da9ec8c4514
|
|
| BLAKE2b-256 |
16da737ab2121355a9d37d2a0814f285b93629c72df2d1bfe6fafb1f6e21bcc5
|
