Technical overview

Platform overview

PhorMind is a multi-tenant SaaS platform for enterprise knowledge retrieval and agent operations. It provides a unified semantic search layer across your organisation’s connected data sources, a governed agent runtime, and a no-code agent builder, all enforcing per-tenant and per-user access controls at query time.

Connector ecosystem

PhorMind ingests and indexes content from Slack, Jira, Confluence, Google Drive, Gmail, Google Calendar, Google Sheets, Google Contacts, SharePoint, GitHub, GitLab, PostgreSQL, MySQL, Elasticsearch, file uploads, websites, email over IMAP, and MCP custom servers. Connectors authenticate via OAuth 2.0 or API key and sync incrementally on a configurable schedule.

REST API overview

The PhorMind API follows RESTful conventions with JSON request and response bodies. Common endpoints include /api/v1/auth/token, /api/v1/workspaces/me, /api/v1/search, /api/v1/chat, /api/v1/connectors, /api/v1/agents, /api/v1/agents/{id}/run, and /api/v1/audit.

Full request and response schemas, authentication flows, and webhook reference are available in the in-platform developer portal after workspace signup.

Security model

• All data in transit is encrypted via TLS.
• Tenant data is logically isolated at the database, vector store, and object store layers.
• Role-based access control is enforced on API routes, and document-level policies restrict which chunks are returned per user.
• Connector OAuth tokens are stored encrypted at rest.
• Agent runs that touch sensitive data sources can require configurable human-in-the-loop approval before execution continues.
• Read and write operations are written to an audit log.