VisionClaw-Agent-Public-Release
VisionClaw-Agent is an open-source multi-tenant AI agent workspace that provides agencies and operators with an always-on AI operations team they own and host themselves.
Ficha actualizada 23 sept 2026, 11:35 · upstream 23 sept 2026, 11:09 · release r135.4-sec
Qué es VisionClaw-Agent-Public-Release
Descripción editorial construida a partir del repositorio, su documentación y sus señales de mantenimiento.
VisionClaw Agent is an open-source, multi-tenant AI agent workspace that enables agencies, operators, and founders to build their own always-on AI operations team. Instead of relying on per-seat SaaS subscriptions, VisionClaw provides a self-hosted platform where you can deploy your own agents with complete ownership of keys, data, and infrastructure.
The platform supports 18 specialized personas working together in real-time to produce deliverables such as research reports, legal documents, financial models, marketing campaigns, slide decks, spreadsheets, and PDFs. Every agent decision is traceable through a governance layer that logs intent, selects tools, and coordinates between specialists. The system integrates with external tools including email, payments, voice assistants, Google Drive, and more — all appearing automatically when needed.
Key features include an AHB safety layer that gates every tool call to prevent destructive operations, tenant-isolated data storage via PostgreSQL, asynchronous orchestration of multiple agents, and a comprehensive tool registry. The platform is designed for agencies needing full control over their AI workforce without vendor lock-in.
- Agencies can deploy an entire multi-agent team (researchers, writers, financial analysts, legal experts) to handle client projects without hiring or managing headcount
- Operators and founders can replace multiple SaaS subscriptions with a single self-hosted platform that provides their own AI workforce
- Solo entrepreneurs can run a complete research, content, sales, and finance team of one agent
- Research teams can use the platform as a testbed for multi-agent orchestration, AHB safety layers, and tool governance
- Legal agencies can have agents analyze contracts across multiple regulatory frameworks simultaneously
- Marketing agencies can run automated research sweeps, content planning, and outreach sequences with full auditability
- Supports 420 total registered tools including file I/O, web scraping, email sending, payments, and document generation
- Provides a live demo at agenticcorporation.net that demonstrates the platform in action
- Offers deterministic artifact authority for consequential code changes requiring server-issued receipts
- Supports 18 specialized personas with roles such as Engineer, Writer, Researcher, Finance Specialist, and Legal Expert
- Every tool call is logged through an AHB safety layer that prevents destructive operations like database pushes or file deletions
- All data is stored in tenant-isolated PostgreSQL instances, ensuring complete data sovereignty
- Autonomous pipelines are disabled by default and require explicit configuration via JURY_AUTOAPPLY environment variable to enable code-mutation fixes
- The platform requires a minimum of one LLM API key (OpenAI or Anthropic) to function, with additional optional keys for email, payments, and voice services
- No warranty is provided — the project is offered AS-IS under its open-source license with no guarantee of uptime or data safety
- The platform does not include built-in customer support or a dedicated help channel; users must rely on community documentation and GitHub issues
- While it supports multiple agents, there is no native multi-tenant database isolation beyond PostgreSQL tenant_id filtering
Capturas del proyecto
Imágenes reales detectadas en la documentación del repositorio.
Perfil de infraestructura
Estimación inicial derivada del repositorio y de los servicios detectados.
Despliegue detectado
Qué señales reales de instalación hemos encontrado.
To deploy VisionClaw-Agent: clone the repository, install dependencies with npm install, set environment variables for DATABASE_URL and SESSION_SECRET (generated via openssl), then run npm run dev to start the server on port 5000. The platform auto-creates its schema and seeds 18 personas plus 41 governance rules upon first initialization. For production deployment, use Docker Compose with a PostgreSQL database, configure your API keys in environment variables, enable the AHB safety layer by setting JURY_AUTOAPPLY=1 if you need autonomous code fixes, and deploy to any supported hosting provider (Heroku, Railway, Render, or self-hosted). The platform is designed for serverless deployment — no Docker containers are required. For production-grade deployments, consider using a managed PostgreSQL instance with proper RBAC and monitoring.
Stack y dependencias
Componentes inferidos a partir de manifests, configuración y documentación.
Puede sustituir a
Relaciones detectadas por el análisis del repositorio y su categoría.
Experiencia real de la comunidad
Ayuda a contrastar la estimación automática con instalaciones reales.
Todavía no hay datos comunitarios para este proyecto.