Help us improve
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
By sancovp
PromptWorld — the meta-compiler *World: a WrightMaster Engineer-CEO that has every skill plus a native subagent per component type (skill / mcp / harness / operating_system / prompt / team / workflow), and 7 PromptGym specialist AIOS agents, that compile / compose / publish prompt systems through the nomicon. Bundles the install-promptworld deploy WIZARD (the agent builds the docker image, boots the container running the Engineer-CEO + PromptGym + the assistant-ui chat / Monaco file explorer / embedded terminal, copies the Claude creds in, and hands the user one URL), compile-a-world (the meta-compiler skill — the nomicon folds on itself), and ingest-into-nomicon-app.
npx claudepluginhub sancovp/sancrev-marketplace --plugin promptworldBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
WHAT: PromptWorld's meta-compiler skill — the architecture patterns + worked examples for building a *World (a sim app = a team-leader + a specialist set + a domain + a driver) such as JobWorld, GameWorld, HealthWorld, or PromptWorld itself, plus how to reuse one pattern, combine them into a new World/app, and compose+publish it over the nomicon. WHEN: when the user (or the CEO/WrightMaster) says 'compile/build/make a World', 'make a new jobworld/gameworld/healthworld', 'what World patterns are there / how is a World built', 'turn this into an app / a sim', 'meta-compiler', 'world compiler', or wants to scaffold a CAVE-fork sim app or fold one into the nomicon (any of).
WHAT: the PromptWorld app-rung — a single flow that ingests an existing Claude-Code system directory (a dir of AIOSes / skills / rules / CLAUDE.mds), converts it into its own {name}-nomicon (the author's framework-of-frameworks holder/search/router), and agentifies that nomicon into a standalone sidecar PromptWorld app (a CEO that carries the nomicon as its world-model, with the ingested AIOS dirs as its specialist agents). It WIRES the already-existing pieces (ingest = copy into the workspace; convert = the skill2framework→fold-into-nomicon pipeline; agentify = PromptWorld scoped to the nomicon) into one drop-dir→get-app flow. WHEN: when the user says 'ingest this claude-code dir', 'turn this system into an app / a webapp', 'make a {name}-nomicon from this', 'convert this dir into its own app', 'app-rung', or wants a Claude-Code directory packaged as a runnable sidecar PromptWorld with an AIOS (any of).
WHAT: the deploy WIZARD for PromptWorld — YOU (the agent) build the docker image and boot a container running the whole system (the Engineer-CEO that has every skill + a native subagent per component type, the 7 PromptGym specialist agents, the assistant-ui chat + Monaco file explorer + an embedded terminal), find the Claude credentials on THIS machine, copy them into the container at boot, and then just hand the user a URL. WHEN: when the user wants to install / run / launch / boot / spin up / deploy PromptWorld, or says 'set up promptworld', 'start promptworld', 'boot the promptworld container' (any of).
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Harness-native ECC plugin for engineering teams - 64 agents, 262 skills, 84 legacy command shims, reusable hooks, rules, MCP conventions, and operator workflows for Claude Code plus adjacent agent harnesses
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Comprehensive startup business analysis with market sizing (TAM/SAM/SOM), financial modeling, team planning, and strategic research
Autopoiesis Plugin - Self-maintaining promise loops for honest AI work
SDNA (Sanctuary DNA) - Gnostic agent workflow DSL with skills for chain development
The doc-mirror claude system: a 1:1 documentation mirror of any codebase, maintained by an invariant doc-mirror STATE MACHINE (per-module doc(m), 6 context/ index files, fork-on-change, no random documents). Bundles doc-mirror-boot (entry + core-loop prime + router), the doc-mirror-{init,seework,change,prompts} state-skills, doc-mirror-install (the host-setup WIZARD), ship-a-plugin (the build+distribute dev-flow), doc-mirror (THE LAW), and make-ai-operating-system (the architect for new systems of this class); the brainhook work-loop + session-start + transition-guard + readonly-guard hooks (plugin-native via CLAUDE_PLUGIN_ROOT); the journal/projects/vision/plan/tracker/commit/cursor + docmirror-search CLI tools; and the doc-mirror law + behavior rules.
CodeNose - Configurable code smell detection for LLMs. Sniffs out duplicate logic, architecture violations, missing logging, and other blind spots.
Instantiate a TWI Jobworld — AI-powered company with agents that run themselves. Enable this plugin and a CEO agent spawns, ready to bootstrap your company.
Uses power tools
Uses Bash, Write, or Edit tools
Share bugs, ideas, or general feedback.
⭐ 0 stars • 🕑 Updated 2026-06-10
📦 Auto-published from the monorepo • CHANGELOG • sancovp/promptworld
PromptWorld is a Claude-Code-native app for building Claude-Code-native software.
You chat with one agent — the Engineer-CEO — and it builds the seven primitive component types that make up any Claude Code system (skills, MCP servers, prompts, harnesses, teams, workflows, operating systems), itself or by delegating to a specialist builder for each type. It runs as a single web app: a chat + a live file workbench + dashboards, served from one container.
PromptWorld is also a worked example of an app built over a "nomicon" — the doc-mirror documentation /
development operating system. Every agent in PromptWorld loads the doc-mirror plugin, so the whole app
carries the ladder for turning a component into an AIOS into a framework and folding it back into the
system. PromptWorld can inspect its own structure and explain, as a living example, how an app like it
is built. It is also the meta-compiler of patterns (the compile-a-world skill): it treats every
*World — JobWorld, GameWorld, HealthWorld, PromptWorld — as a named architecture pattern (a *World = a
sim: a team-leader + specialists + a domain + a driver), so an agent can reuse a pattern or combine them
into a new World/app and fold it up the nomicon ladder. (*World is a CAVE-fork resemblance, a way of
doing this — not a rigid type; there is no push-button general generator yet, by design.)
PromptWorld ships as a Docker image built from the monorepo. From the app directory:
cd application/promptworld
# 1. Build the image (stages promptworld + the doc-mirror plugin + cave + sdna, builds the SPA in-layer)
bash deploy/build.sh build
# 2. Boot a container on host port 3858 (mind_of_god-safe: create -> docker cp creds -> start)
# REQUIRES the provider key in your env so the agents can run:
export MINIMAX_API_KEY=... # the MiniMax-M3 key the agents authenticate with
bash deploy/build.sh boot 3858
# 3. Smoke-test it end-to-end from inside the container (health, SPA, a real CEO reply, a subagent build)
bash deploy/build.sh e2e
# stop
bash deploy/build.sh stop
Then open http://localhost:3858.
Auth model. The agents run on MiniMax-M3 (Minimax-M3[1m], base URL https://api.minimax.io/anthropic).
The key is passed into the container at boot as MINIMAX_API_KEY and exported as ANTHROPIC_AUTH_TOKEN.
A Claude subscription credential is also copied in (for the embedded terminal), but the agent turns use
the MiniMax provider token, not OAuth.
PromptWorld's World module is the guild that builds things:
agents/engineer-ceo.md) — the single agent you chat with. It has every skill and
one native subagent per component type, so it can build anything itself or delegate.promptgym/<craft>/, one per component type. Each is a directory
that codes an agent (its own CLAUDE.md + .claude/rules + .claude/agents), and each specialist
works ONLY in its own directory.| craft | builds | master |
|---|---|---|
skill | a SKILL (lean SKILL.md + resources/) | Skillwright |
mcp | an MCP server (FastMCP tools) | Toolwright |
prompt | a prompt / persona (a "guy" in a role) | Promptwright |
harness | a harness (sequenced typed outputs across a turn) | Harnesswright |
team | a team / subagent set on a shared task list | Teamwright |
workflow | a deterministic fan-out script (the Workflow tool) | Flowwright |
operating_system | an AI OS (a directory + core loop that governs) | Systemwright |
The shared, re-skinnable world description lives in promptgym/world-context.md; the shared global agent
context in promptgym/global-context.md.
Every PromptWorld agent loads the doc-mirror plugin, which carries the rungs:
component -> AIOS -> framework -> fold-into-nomicon -> app
as real skills/CLIs (make-ai-operating-system, the skill2framework prompt-tree, nomicon-atomize,
ship-a-plugin, and the app-rung skill). Ask the CEO "how is PromptWorld built / what is the nomicon"
and it inspects its own dirs and explains itself as the example.
Five pages, selected from the left panel:
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claim