What You Can Do with Hrida.ai
One interface for every AI model. Private, extensible, and built for teams.
Hrida.ai replaces the patchwork of AI tools your team juggles daily - ChatGPT for writing, a separate app for image generation, another for document search, spreadsheets full of prompts nobody can find. Everything lives in one place: conversations, knowledge, tools, and the models that power them.
Department-scoped AI agents with their own knowledge, role, and access control.
Configure HR, IT, Documentation, Legal, and Sales agents from the admin panel — each with its own knowledge base (populated from your docs), scoped system prompt, introduction, and escalation path. Access control restricts each agent to the right users and groups, no code required.
| Multiple agents | HR, IT, Documentation, and more — each fully isolated with separate KB and config |
| Per-agent knowledge | Each agent indexes its own docs folder or live website via sitemap |
| Role-scoped behavior | Introduction, scope, and escalation contact configured per agent |
| Access control | Restrict each agent to specific users and groups — applies instantly, no re-setup |
| One-click bootstrap | 4-step SSE stream creates KB, filter, tool server, and agent model automatically |
| Incremental sync | Add new docs without rebuilding — only new files are uploaded |
Video meetings, AI notes, scheduling, and agent rooms — natively in Studio.
A plugin-based meeting suite that integrates Jitsi Meet (self-hosted video with SSO), Whisper + Ollama (automatic transcription and meeting summaries), scheduling synced with the built-in Calendar, and multi-agent discussion rooms — all without leaving Hrida.ai.
| Video conferencing | Jitsi Meet or Zoom with JWT SSO — users auto-authenticate from their Hrida.ai session |
| AI meeting notes | Upload recording audio → Whisper transcription → Ollama summary + action items |
| Scheduling | Book meetings directly inside Studio; syncs natively with the built-in Calendar as a CalendarEvent — no external service |
| Multi-agent rooms | Facilitator → Analyst → Summarizer agent chain collaborates on any topic in a shared channel |
| Plugin-based | Each capability is a configurable plugin — enable what you need, skip what you don't |
A collaborative Kanban task board, built into Hrida.ai.
Create boards, assign work, and track status without leaving Studio. Turn a Channel discussion into a tracked task with one slash command, and follow every claim, delegate, and status change in a full activity trail.
| Boards | Named boards, public or restricted, each with independent membership |
| Member roles | Viewer, Editor, and Admin roles on restricted boards |
| Task lifecycle | Open → Assigned → In Progress → Review → Done, with priorities from low to urgent |
| Claim & delegate | Self-assign open tasks or reassign them to a teammate |
| Live updates | Board and task changes stream in real time over SSE |
| Channel integration | Create a task straight from a Channel message with /task board:<name> |
Design, test, and publish multi-step AI workflows on a visual canvas.
Wire together AI agents, logic gates, data transforms, and tool calls into a graph that runs as a repeatable workflow. Generate entire workflows from a plain-English description with AI Assist, or get intelligent suggestions for the next nodes to add. Ship to production with a built-in review and publish lifecycle.
| Visual canvas | Drag-and-drop node palette with 13 node types across Core, Tools, Logic, and Data categories |
| AI Assist | Generate a complete workflow from a description, or get context-aware node suggestions for an in-progress workflow |
| Lifecycle management | Draft → Stage for Review → Approve/Reject → Publish with role-based controls |
| Versioning | Snapshot, rollback, and diff any two versions of the workflow graph |
| ▶️ Live run panel | Execute and stream SSE output in real time from within the editor |
| Human-in-the-loop | Pause runs at any node for inbox-based human approval (approve / modify / reject / delegate) |
| Inbox | Per-user task queue for pending review decisions; real-time badge updates via SSE |
| Review Queue | Admin queue for approving or rejecting staged workflows; optional multi-stage approval matrix |
| Test cases | Save named input fixtures and replay them to verify workflow behavior after edits |
Import, govern, and publish REST APIs as versioned, auditable tools your agents can call.
Import an OpenAPI spec, configure upstream auth, and take it through a review lifecycle modeled on Agent Builder's own draft → staged → published flow. Publishing an API automatically registers it as a Tool Server — instantly callable by any Agent Builder agent, with a full audit trail and one-click rollback.
| Spec import | Create a draft API definition from an OpenAPI 3.x or Swagger 2.0 spec in JSON or YAML |
| Governed lifecycle | Draft → Staged → Published, with reject, revise, and rollback |
| Audit log & version history | Every lifecycle transition and every published version is recorded and snapshotted |
| AI Message Mappers | Reshape upstream JSON with an AI-assisted, sandboxed expression — plus a built-in test runner |
| AI Protocol Mappers | Convert JSON ⇄ XML/SOAP/form-urlencoded, auto-detected from a sample |
| Auto tool-server registration | Publishing an API instantly makes it a callable tool for Agent Builder agents |
| Built-in API gateway | Every call goes through the gateway: rate limits, size caps, signed caller identity, request validation, response masking and a per-call audit log |
| Enterprise upstream auth | Bearer, API key, Basic, OAuth2 client credentials, custom CA and mutual TLS — secrets encrypted and rotatable |
| API Spaces & RBAC | api_developer, lifecycle_manager, space_admin roles scoped per space |
Generate Documentation, Research, Sheets, and PPT Slides from a prompt — or start from a template and edit it.
A standalone workspace for creating persisted content, separate from chat: describe what you want and let a model draft it, or start from a ready-made template and edit it in a dedicated editor per format.
| Four formats | PPT Slides, Documentation, Research, and Sheets, each with its own editor |
| Generate from a prompt | Pick a model and describe what you want — get a complete first draft |
| Ready-made templates | Pitch Deck, SOP, Market Research, Budget Tracker, and more — or start blank |
| Autosave | Edits save automatically — no explicit save step |
Talk to any model, switch mid-conversation, and keep your context intact.
Your conversations are the core of Hrida.ai. Chat with Ollama, OpenAI, Anthropic, or any OpenAI-compatible provider from a single interface. Attach files, search the web, execute code, and let the AI use tools - all without leaving the chat.
| Multi-model chats | Run two models side-by-side and compare responses |
| File & image uploads | Attach documents, images, and code for the AI to analyze |
| Web search | AI searches the web and cites sources in real time |
| Code run | Run Python directly in the browser or via Hrida Terminal |
| Message queue | Keep typing while the AI responds - messages send automatically |
| Memory | The AI remembers facts about you across conversations |
| Folders, tags, pins | Organize conversations however you work |
| Voice & audio | Speech-to-text, text-to-speech, hands-free voice & video calls |
| Image generation | Create and edit images with DALL-E, Gemini, ComfyUI, and more |
| Automations | Schedule prompts to run automatically on recurring schedules |
| Task management | Models maintain structured task lists for multi-step workflows |
Give your AI access to your documents - and let it find what matters.
Upload files, build knowledge bases, and let the AI retrieve exactly the information it needs. Choose between vector search (RAG) for large collections or full-content injection for precision. With native function calling, models autonomously search, browse, and synthesize across your entire knowledge base.
| 9 vector databases | ChromaDB and PGVector (officially maintained), plus additional options: Qdrant, Milvus, Elasticsearch, and more |
| Hybrid search | BM25 + vector search with cross-encoder reranking |
| 5 extraction engines | Tika, Docling, Azure, Mistral OCR, custom loaders |
| Agentic retrieval | Models search and read your documents autonomously |
| Full context mode | Inject entire documents - no chunking, no guessing |
Wrap any model with custom instructions, tools, and knowledge to build specialized agents.
Create a "Python Tutor" that always uses your style guide. A "Meeting Summarizer" with your company's template. A "Code Reviewer" with your linting rules baked in. Every agent is a configuration wrapper - pick any base model, bind knowledge, tools, and a system prompt.
| Model presets | System prompts, tools, knowledge, and parameters in one package |
| Dynamic variables | {{ USER_NAME }}, {{ CURRENT_DATE }} injected automatically |
| Bound tools | Force-enable specific tools per model |
| Access control | Restrict models to specific users or groups |
| Global defaults | Set baseline capabilities and parameters for all models |
Group related chats, knowledge, and instructions into a named workspace for a specific goal.
A project is a persistent workspace with its own memory: write instructions once and every conversation inside the project inherits them. Share with teammates via access grants and pick up exactly where you left off.
| Custom instructions | Free-form context injected into every conversation inside the project |
| Access grants | Share with specific users or groups with read or write permissions |
| Rich metadata | Attach arbitrary JSON data and meta fields for custom integrations |
| Description | Human-readable summary shown in the paginated project list |
Partition your instance into isolated AI environments for different business units, clients, or teams.
A Catalog is a top-level isolation boundary with its own LLM providers and API keys. Spaces are collaborative areas inside a catalog — each with its own members, roles, and optional LLM overrides. Run one instance that simultaneously serves multiple departments without any cross-contamination.
| Catalog isolation | Separate LLM providers, API keys (encrypted at rest), and access rules per catalog |
| Space roles | Viewer / Editor / Admin roles per space with member management |
| LLM overrides | Per-space model and provider override takes precedence over catalog defaults |
| Encrypted keys | LLM API keys stored Fernet-encrypted; never returned in plaintext via API |
Connect your inbox and let AI triage every message.
The Email Intelligence Plugin links Hrida.ai to any IMAP/SMTP inbox. AI reads incoming mail, assigns urgency scores, auto-tags by topic, flags spam, drafts replies, and cross-references meeting invites with your CalDAV calendar.
| Urgency scoring | Each message rated 1–5 so you know what to read first |
| Auto-tagging | Topic tags applied automatically (billing, legal, internal, urgent…) |
| Spam detection | Spam score 0–1; messages above the threshold are moved to Spam automatically |
| Draft replies | Ready-to-send AI-generated reply for each triaged message |
| CalDAV awareness | Meeting invites cross-referenced with your calendar for scheduling conflicts |
| Encrypted credentials | IMAP / SMTP / CalDAV passwords stored Fernet-encrypted; never returned in API responses |
Write, think, and refine with AI by your side.
A dedicated workspace for content that lives outside of any single conversation. Draft with a rich editor, use AI to rewrite text in place, and attach notes to any chat for precise context injection - no chunking, no vector search.
| Rich editor | Markdown and Rich Text with a floating formatting toolbar |
| AI Enhance | Rewrite or improve selected text in place |
| Context injection | Attach notes to any chat for full-fidelity context |
| Agentic access | Models can search, read, and update notes autonomously |
Where your team and AI think together, in real time.
Persistent, shared spaces where humans and AI models participate in the same conversation. Tag @gpt-4o to draft a plan, then tag @claude to critique it - your whole team sees both responses in one timeline.
| @model tagging | Summon any AI model into the conversation on demand |
| Threads & reactions | Replies, pins, and emoji reactions |
| Access control | Public, private, group-based, and direct message channels |
| AI channel awareness | Models search and synthesize across channels autonomously |
Give your AI a real computer to work on.
Connect a real computing environment to Hrida.ai. The AI writes code, executes it, reads the output, fixes errors, and iterates - all from the chat. Handles files, installs packages, runs servers, and returns results.
| Code run | Runs real commands and returns output |
| File browser | Browse, upload, download, and edit files in the sidebar |
| Website preview | Live preview of web projects inside Hrida.ai |
| Isolation optional | Docker container or bare metal |
Add any capability with Python tools, pipelines, MCP, or OpenAPI servers.
Hrida.ai is a platform, not a locked-down product. Write Python tools that run inside the chat. Connect external services via OpenAPI or MCP. Build pipelines that filter, transform, or route every message. Import third-party extensions with one click.
| Tools & Functions | Python scripts that run in the chat - built-in code editor included |
| Pipelines | Modular plugin framework for filters, providers, and custom logic |
| MCP support | Native Streamable HTTP for Model Context Protocol servers |
| OpenAPI servers | Auto-discover tools from any OpenAPI-compatible endpoint |
| Skills | Markdown instruction sets that teach models how to approach tasks |
| Prompts | Slash-command templates with typed input variables and versioning |
Control who gets in, what they can do, and how it connects to your identity stack.
Hrida.ai is multi-user from day one. Define roles, create user groups, set per-model access, and integrate with your identity provider. From a solo install to an organization with thousands of seats.
| RBAC | Roles, groups, and per-resource permissions |
| SSO/OIDC/LDAP | Federated authentication with any identity provider |
| SCIM 2.0 | Automated user and group provisioning |
| API Keys | Programmatic access for scripts, bots, and integrations |
Monitor usage, evaluate models, enforce budgets, and communicate with your users.
| Analytics | Usage dashboards with message volume, token consumption, and cost tracking |
| Evaluation | Model arena, A/B testing, and ELO-based leaderboards |
| Model Pricing | Configure per-token USD costs so analytics and usage limits can track spend |
| Usage Limits | Per-user or per-group daily/monthly token and cost quotas with hard or soft enforcement |
| Banners | Customizable system-wide announcements |
| Webhooks | Notifications for sign-ups, chat completion, and external integrations |
| Incident Management | Plugin-based integration with Grafana OnCall, PagerDuty, and Opsgenie |
Docker, Kubernetes, pip, bare metal - with horizontal scaling and production observability.
| Docker & Compose | One-command deploy with GPU support |
| Kubernetes & Helm | Production-ready orchestration |
| pip install | pip install hrida-ai-studio && hrida-ai-studio serve |
| Cloud storage | S3, GCS, Azure Blob for stateless instances |
| OpenTelemetry | Traces, metrics, and logs to your observability stack |
| Horizontal scaling | Redis-backed sessions, multi-worker, multi-node |