features
Everything in Pro.
220 agents. Six workbenches. Agent Studio. Local-first runtime. BYOK across every major LLM. $29/month gets you all of it.
the workbenches
Six workbenches you can't get elsewhere.
Pre-built multi-agent pipelines for the work most people actually want done. All bundled in Pro.
Coding Forge
Production-grade code generation, on demand
A multi-agent pipeline that produces working code. Architect drafts the structure, generator writes the code, critic reviews against your spec, tester runs the tests. You get a downloadable artifact with the code + the test suite + the receipts.
- Architect → Generator → Critic → Tester pipeline
- Real test execution before delivery
- Up to 50 builds per month
- Artifact downloads (zip + tests), 30-day retention
- Submit from dashboard OR from your local agent
Video Studio
Script-to-render in one pipeline
Hand it a topic or a script. It writes shot lists, generates clips, narrates with a voice you pick, scores with music, and assembles everything into a 720p export. Bring your own Sora/Veo/ElevenLabs keys.
- Scriptwriting + shot list + clip generation
- Voice narration with style options
- Music score + final assembly
- 720p export with thumbnails
- 20 video projects per month
Coherence Cloud
Frontier-model critic for your agents
Producer models can't grade themselves well. Coherence Cloud routes your agent outputs to a frontier judge (Claude Opus, GPT-4.x) that catches drift, hallucinations, and weak self-evals you'd otherwise miss.
- Catches drift and weak self-evals
- Architecture sanity-check judge
- 5-axis coherence verdicts with reasoning
- 1,000 critic + judge calls per month
- Drop-in upgrade over the free local critic
Swarm Mode
Orchestrated multi-agent delegation
A Commander agent decomposes your task into distinct sub-tasks, delegates one to each specialist, then a Synthesizer agent merges the results into a unified answer. Not a parallel race — a real orchestration matching the OSS SwarmCommander pattern.
- Commander decomposes → delegates → synthesizes
- Live grid of each member agent's progress
- Pick member agents or let the Commander choose
- Synthesized unified answer at the end
- 2-20 agents per swarm
Nexus Hub
Your local Nexus, embedded in the cloud dashboard
The Nexus Hub embeds your actual local Nexus dashboard pages through the runner WebSocket — not a re-implementation, the real thing. When your runner is online, you see what you'd see locally.
- Direct embed via HTTP-over-WebSocket tunnel
- Available when your local runner is online
- Sandboxed iframe for safety
- Reload + open-in-tab controls
- No duplicate state — what you see is what's there
Agent Studio
Build your own roster-quality agents
Two builders in one workbench. Brain Builder generates a persistent companion agent (the 7-file openclaw bundle). Specialist Builder generates a stateless roster-style agent matching the 3-layer prompt architecture of the 220 marketplace specialists.
- Brain Builder: full openclaw 7-file bundle
- Specialist Builder: ROLE_INSTRUCTIONS for the 3-layer system
- Downloadable .json bundles ready to deploy
- Live tabbed preview of every generated file
- Same prompt architecture as the 220-agent roster
the bigger picture
The repryntt system itself.
Workbenches are the front of the house. This is what they sit on top of — the runtime, the agents, the philosophy.
Local-first runtime
The full repryntt framework runs on your hardware — laptop, server, Jetson, even Android via Termux. Agent memory always lives on your device. The cloud subscription is convenience, not lock-in.
220 specialist agents
158 marketplace business roles + 62 frontier-science specialists across 33 departments. Each carries a tuned persona, tool policy, and workflow profile. Not blank chats — colleagues that already know the job.
Three-layer prompt architecture
PROFESSIONAL_BASE (the agent loop + hard constraints) + DEPT_FRAMEWORK + ROLE_INSTRUCTIONS. Every specialist passes through UNDERSTAND → PLAN → EXECUTE → VERIFY → DELIVER. No fabrication, no unverified code, no untraceable claims.
380 built-in tools, growing weekly
Knowledge & web search, file & code, math & stats, browser, memory, swarm dispatch, creative writing — the full tool registry every agent can call. 380 tools across 35+ categories today; new tool drops ship on a weekly/monthly cadence. Tools chain end-to-end: research → process → verify → present.
BYOK across providers
Connect OpenAI, Anthropic, xAI, Google Gemini, NVIDIA NIM, or local Ollama/vLLM. The router picks per task. You pay the model provider directly — we never mark up tokens.
Local Runner ↔ Cloud pairing
Pair your local OSS install with the cloud dashboard via a runner token. Jobs submitted from the SaaS execute on YOUR machine with YOUR keys. The Nexus Hub embeds your local pages directly.
Robotics-ready (OSS)
On Jetson Orin, the installer offers the full robotics stack: ROS2 bridges, LeRobot integration, tiered vision pipeline (local SmolVLM → NVIDIA NIM fallback). Embodied agents on your hardware, today.
Open core, MIT-licensed
The framework is open source under the MIT license. Self-host the whole thing for free, fork it, ship it inside your product. Pay $29/mo when you want the hosted dashboard, the cloud workbenches, and email support.
the platform
Everything around the agents.
Dashboard, API, routing, support — the things that make hosted worth $29/month over self-hosting.
Hosted dashboard
Works in any browser. No Jetson, no Linux, no Docker. Submit agent jobs, view results, manage settings — all from the web.
REPRYNTT_API_KEY
Call the platform from your own code. Drop it into a Python script, a CI pipeline, a Zapier-like flow — anywhere you'd hit an API.
Multi-LLM routing
Connect OpenAI + Anthropic + xAI + Google keys. The platform routes each task to the right model — cheap models for cheap work, frontier for hard.
Brain Pulse activity
Live monthly usage, runner status, active jobs, recent feed. The dashboard heartbeat — see what your agents are doing in real time.
Email support
hello@ai158z.com — 1 business day response. Founder-direct for the first 100 customers.
the roster
220 agents across 33 departments.
Tier 1 — The Marketplace 158
The original 158-agent marketplace roster: every business role from Strategic Research Analyst to DevOps Engineer to Investigative Journalist. Each has a tuned persona, tool policy, and workflow profile.
Tier 1b — The Scientific 62
62 frontier-research specialists across 12 departments: Plasma Physicists, Fusion Engineers, Materials Scientists, AI Researchers, Aerospace Engineers, Climate Modelers, Biotech Researchers, Cryptographers, and more.
Tier 2 — System extensions
Agents that ship with workbenches: Forge architect/generator/critic/tester, Video producer/scripter/scorer, Coherence judge, Swarm Commander + Synthesizer. These extend the system rather than counting toward the OG roster.
Tier 3 — Your custom agents
Built today via Agent Studio. Brain Builder gives you a persistent companion (openclaw 7-file bundle). Specialist Builder gives you a roster-style stateless specialist using the same 3-layer prompt architecture.
what's next
On the roadmap.
Things we're building, in priority order.
v1.1
Worker — hourly autonomous agent
Spin up a 24/7 autonomous agent worker. Watches your repo / inbox / data. Billed by the hour while running, pause anytime.
v1.2
Managed LLM access
Don't want to manage your own API keys? Optional add-on where we handle the OpenAI/Anthropic relationship and bill you with the platform fee.
v2
Multi-user teams + shared workspaces
Share agents and workflows across a small team. Permission scopes. Centralized billing.
v2
Compute marketplace
Community providers run verified inference capacity. Lower model costs at scale. Stripe Connect payouts.
v2
Robotics Cloud (managed)
Managed deployment of the OSS robotics stack — pre-paired Jetson units, fleet coordination, remote teleop.