Python 3.14+ · BUSL 1.1 · Source Available
An autonomous product studio you operate.
Describe a product to build, or hand it an existing codebase: a synthetic organisation of role-based AI agents, with budgets, memory, and governance, is designed to plan and deliver it. The platform is built and tested; the agent runtime that runs the organisation is in active development, in the open.
Honest status: The platform, infrastructure, and subsystem libraries are built and tested. The autonomous agent runtime that makes the organisation execute work end to end is in active development and tracked in the open. Capability sections on this page are marked as available now or in active development. Roadmap · Issue tracker · Get in touch
How It Is Designed To Work
Steps 1-2 (pick a template, configure via wizard or REST API) work today. Steps 3-4 illustrate the designed runtime behaviour, agents collaborating and the operations view, which is in active development.
Choose a template
Solo Builder 2 agents 2 depts
Tech Startup 5 agents 3 depts
Engineering Squad 8 agents 3 depts
Product Studio 10 agents 5 depts
▶ Agency 12 agents 5 depts
Research Lab 7 agents 3 depts
Enterprise Org 30+ agents 9 depts
Or define your own with YAML config.Built-in templates from 2 to 30+ agents. Fully customisable roles, personalities, budgets, and tools.
POST /api/v1/setup/company
template: "agency"
company_name: "Acme AI Lab"
201 Created
agents_created: 12
departments:
- Operations (PM, Scrum Master)
- Engineering (3 developers)
- Design (UX, UI)
- QA (QA Engineer)
- Creative (Writer, Strategist)Setup wizard or REST API. Auto-assigns LLM models, generates agent names, configures tool access.
[TASK] #12 created: "Redesign checkout flow"
[ENGINE] PM decomposed #12 into 4 subtasks
[ENGINE] #12a -> UX Designer (role-based)
[ENGINE] #12b -> Backend Dev (load-balanced)
[ENGINE] #12c -> Frontend Dev (role-based)
[ENGINE] #12d -> QA Engineer (role-based)
[COMMS] UX Designer -> PM: "Wireframes ready"
[DONE] #12a approved by PM (quality: 94%)
[COMMS] Meeting: sprint review (round-robin)Designed runtime behaviour (in active development): DAG-based decomposition, routing strategies, meetings, delegation, and conflict resolution.
Agents active: 12 across 5 departments
Tasks completed: 47
Total cost: EUR 18.92
Coordination overhead: 14% (healthy)
Budget remaining: EUR 81.08 / 100.00
Top Performers:
Backend Dev quality: 95% tasks: 11 cost: EUR 3.20
UX Designer quality: 93% tasks: 8 cost: EUR 2.15
PM quality: 91% tasks: 9 cost: EUR 4.80Designed operations view (in active development): per-token cost tracking, coordination overhead, performance scoring, budget enforcement.
What Makes SynthOrg Different
Not just agent orchestration. SynthOrg is designed as a complete organisational system: HR, governance, personality-driven teams, and cost controls, modelled as a real company rather than a chain of function calls.
Orchestration Engine
Routing strategies and execution loops, designed as one engine.
Designed for auction, role-based, and load-balanced task routing, DAG decomposition, conflict resolution, and crash recovery. Built and unit-tested as components; running it end to end is the runtime work in active development.
Sequential, parallel, Kanban, and Agile sprint workflows in the design.
Governance & Security
Every action classified, audited, and gated, by design.
A designed hybrid SecOps model: a fail-closed rule-engine fast-path plus an LLM evaluator for uncertain cases, four autonomy tiers (full, semi, supervised, locked), and budget enforcement. Implemented and unit-tested; enforced on a live run once the agent runtime is live.
Progressive trust: privileges never auto-granted, auto-downgrade on violations.
Agent Identity
Agents designed with personalities, careers, and reviews.
Big Five personality dimensions for team compatibility, seniority levels from Intern to C-Suite, quality scoring, and hiring/promotion/offboarding with memory archival. Modelled and tested as components.
Not interchangeable workers; agents are designed with identity and growth.
Memory & Knowledge
Multiple memory types. Knowledge designed to outlast any agent.
Working, episodic, semantic, procedural, and social memory with hybrid prompt and retrieval injection, designed to survive offboarding and transfer to successors. The pipeline runs once a live agent exercises it (in active development).
Organisational knowledge designed to persist across the agent lifecycle.
Architecture
The designed, modular, protocol-driven architecture. Every component is pluggable and built as tested code; the agent runtime that drives it end to end is in active development.
Security-First by Design
Autonomous agents need guardrails. The CI/CD pipeline and container infrastructure security below is in place today; the application-runtime guardrails are designed and unit-tested, and become active with the agent runtime (in active development).
Runtime Protection (in active development)
Designed hybrid SecOps: a fail-closed rule-engine fast-path with an LLM evaluator for uncertain cases, classified action types, and four autonomy tiers (full, semi, supervised, locked). Implemented and unit-tested as components; enforced on a live run once the agent runtime lands.
- Fail-closed rule engine + LLM fallback evaluator
- Four autonomy tiers with progressive trust
- Output scanning with redaction policies
- Persistent audit trail for every decision
Hardened Infrastructure
Distroless containers, digest-pinned images, and CIS Docker Benchmark compliance. Every image is scanned by two tools and signed before push.
- Chainguard distroless runtime (no shell)
- Trivy + Grype dual scanning
- cosign keyless image signing
- CIS 5.3 / 5.12 / 5.25 compliance
Continuous Security
Eight security scanners run across the full development lifecycle, from first commit to production deployment. Secret detection, SAST, DAST, dependency auditing, license compliance, and supply chain scoring.
- gitleaks, CodeQL, ZAP DAST, zizmor
- pip-audit + npm audit + Socket.dev
- OSSF Scorecard supply chain maturity
- Signed commits + required reviews
Use Cases
The studio is designed for organisations like these. They describe the target shape of a SynthOrg company, not validated end-to-end runs; the agent runtime that executes them is in active development.
AI Dev Team
CTO + Engineers + QA. Autonomous software development with code review, testing, and deployment pipelines.
- DAG-based task decomposition
- Git tools + code sandbox
- Three-tier quality scoring
Research Organization
Director + Analysts + Reviewers. Literature review, data analysis, and collaborative report generation.
- Memory retrieval pipeline
- Hybrid search (dense + BM25)
- Automated consolidation
Content Pipeline
Editor + Writers + Fact-checkers. Content creation with editorial oversight, review cycles, and quality control.
- Meeting-based coordination
- Conflict resolution protocols
- Approval gate workflows
Customer Support
Manager + Agents + Escalation. Tiered handling with knowledge sharing and automatic performance tracking.
- Hierarchical delegation
- Shared knowledge base
- Performance scoring
Security Operations
SecOps lead + Analysts + Incident responders. Automated scanning, threat assessment, and coordinated response.
- Progressive trust levels
- Persistent audit trails
- Sandboxed tool execution
Data Analytics
Lead Analyst + Data Engineers + Visualization specialists. End-to-end analysis pipelines with budget tracking.
- Multi-project support
- Automated reporting
- Per-token cost tracking
See the Dashboard
The React 19 dashboard for agents, tasks, budgets, and communications, with live WebSocket updates. Shipped today; it becomes a live operations view as the agent runtime comes online.
License & Usage
SynthOrg uses the Business Source License 1.1: source available, production-friendly for most users, converting to Apache 2.0 automatically.
Always Free
- Personal learning and experimentation
- Academic and research use
- Development, testing, and evaluation
- Contributing to SynthOrg itself
Free in Production
- Organizations with <500 employees and contractors (including affiliates)
- Non-competing use: internal tools, automation, or products where SynthOrg is not the core value
- No separate license needed; see LICENSE for full terms
Commercial License
- Organizations with 500+ employees
- Offering SynthOrg as a hosted service
- Reselling or embedding as core product
- May be granted at no cost; just ask
Every version converts to Apache 2.0 automatically. We welcome feedback on the license: start a discussion or open an issue.
Read the full licensing rationale →Operate a company that builds.
Get StartedGet in Touch
Questions about licensing, commercial use, or the framework? We'd love to hear from you.
Message sent
Thanks for reaching out! We'll get back to you soon.