DronaHQ
Also known as: DronaHQ, dronahq.com, dronahq.ai, DronaHQ Agents, DronaHQ Agentic Platform, Deltecs Infotech
DronaHQ, a low code vendor, offers an agentic platform whose agents choose from 1,000+ tools and MCP servers, start on webhooks, schedules and 145+ app events, and are tested with scored datasets and twelve reviewer types on a managed cloud.
DronaHQ began as a low code builder for internal tools and mobile apps and now sells a separate agentic platform alongside it. An agent is configured with instructions, a model chosen from several providers, and tools: prebuilt integrations to services such as Gmail, Slack, GitHub and Salesforce, connector queries against the customer's databases and APIs, Python or JavaScript code, the customer's own MCP servers, and inbuilt tools such as web search. The agent decides which tools to call and in what order, and a knowledge base adds retrieval over uploaded documents and crawled URLs.
Agents start from many directions. Each agent can carry several triggers at once: a webhook, a scheduler, chat, WhatsApp, Instagram and Facebook Messenger, and 145+ platform triggers that fire on events in Gmail, Slack, GitHub, Jira, Salesforce, HubSpot, Stripe and similar systems. Published agents run as a standalone chatbot, embed in a DronaHQ app or a website, or slot into an automation.
For testing, datasets hold test cases that specify context, the tool the agent must call and a reviewer for the final answer. Twelve reviewer types range from exact match and JSON schema checks to factuality and custom LLM judges, and evaluation runs report average scores that can be compared side by side. Every run is traced with each tool call, its parameters and response, token usage and credit consumption.
Agents run only on DronaHQ's managed cloud. Models run on DronaHQ's own key, and customers cannot bring their own. Agent to agent and automation tools are marked coming soon, and long term memory is announced for a later release. SSO, RBAC and SOC 2 sit on the Enterprise column of the agent pricing table. A catalog of installable agents covers onboarding, support triage, appointment booking, recruitment, event promotion and KYC.
Vendor details
Canonical URL
https://www.dronahq.com
Category
Agent builder
Subcategory
Agentic platform from a low code app builder vendor
Funding status
DronaHQ is private, operated by Deltecs Infotech Pvt Ltd. No funding rounds are published on its own pages.
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
Agent tools come from a library of 1,000+ prebuilt integrations (Gmail, GitHub, Slack, Notion, Google Sheets, Salesforce, HubSpot), connectors and connector queries configured in the account, Python or JavaScript code, the customer's MCP servers, and inbuilt web search. 145+ platform triggers plus webhook, scheduler, WhatsApp, Instagram and Facebook Messenger start agents. Models from OpenAI, Anthropic, Gemini, DeepSeek, Cohere and xAI run through a built in DronaHQ key.
In practice
A support team installs the support triage agent, adds its ticketing connector and a WhatsApp trigger, and lets unresolved cases escalate to staff.
An operations group schedules a nightly reporting agent and watches each tool call and its credit cost in the trace view.
A builder creates a dataset of refund requests with expected tool calls and a factuality reviewer, then compares eval scores before and after a prompt change.
Sources & related URLs
Related / legacy domains
Research sources
Agentic Index coverage score
11.0 / 14 capabilities · 79%
| Integrations & Tool Calling | Full |
|---|---|
|
Prebuilt DronaHQ Tools connect authenticated accounts for Gmail, Slack, Google Calendar, GitHub, Notion, Apollo and more, connector queries reach the customer's databases and APIs, code tools run Python or JavaScript, and the customer's own MCP servers can be attached. The agent chooses which tools to use and in what order. Sourcedocs.dronahq.com/agents/getting-started/tools-overviewread 2026-10-02 |
|
| Workflow Orchestration | Partial |
|
A single agent follows multi step plans across its tools within a set iteration count, and published agents can be called from DronaHQ automations as a step. Agents cannot yet call other agents, because the Agent and Automations tool types are still marked coming soon, so the multi agent orchestration on the marketing page is not yet available. Sourcedocs.dronahq.com/agents/getting-started/introductionread 2026-10-02 |
|
| Knowledge Grounding & RAG | Full |
|
A knowledge base of documents and crawled URLs, with crawl settings, is attached to an agent for retrieval over organization specific content. Knowledge base operations include document embedding and vector search, metered in credits. Sourcedronahq.com/agents-faqread 2026-10-02 |
|
| Human Oversight & Guardrails | Partial |
|
Customers set guardrails with prompt and output policies, tool limits per agent, and keywords that scope instructions to the current user. DronaHQ advertises human in the loop support, and the KYC template uses guided confirmation steps, but no step where a person approves an agent action before it commits is described. Sourcedocs.dronahq.com/agents/getting-started/introductionread 2026-10-02 |
|
| Security, Identity & Governance | Full |
|
Access controls include scoped permissions, role based access control and version control, and SSO is offered on the agent Enterprise plan and on the app platform as SAML or OpenID Connect. DronaHQ is SOC 2 Type II and ISO 27001 certified and audited for GDPR, with HIPAA as an agent add on. Every agent deployment is said to include SSO and role based access, though the agent pricing table shows SSO, RBAC and SOC 2 only for Enterprise. Sourcedronahq.com/pricingread 2026-10-02 |
|
| Observability & Auditability | Full |
|
Every agent run is recorded with a step by step breakdown of each tool invoked, its parameters and responses, alongside execution logs, tool call traces, token usage and credit consumption, error monitoring and developer request logs. Tracing is kept for 30 days on Starter and 90 on Business. Sourcedronahq.com/agents-faqread 2026-10-02 |
|
| Memory & State Persistence | Partial |
|
The Starter plan includes short term memory, which stores intermediate results, context or conversation history within a session. Long term memory is listed as coming soon on the Business plan, so there is no memory across conversations with a set scope and lifetime yet. Sourcedronahq.com/agents/pricingread 2026-10-02 |
|
| Deployment & Data Residency | Partial |
|
DronaHQ says agent deployments come with data residency controls, but agents currently run only on DronaHQ's cloud, with no region list, customer environment or way to choose one. Zero data retention and 60 day retention add ons cover data handling, not residency, and while the DronaHQ app builder can be self hosted, the agent platform cannot. Sourcedronahq.com/agents/templatesread 2026-10-02 |
|
| Prebuilt Agents, Templates & Packs | Full |
|
Installable prebuilt agents each do their own job, among them an employee onboarding agent, a support triage agent, a WhatsApp appointment agent, a recruitment agent, an event promotion agent and a KYC agent, each with an Install agent link and grouped under Sales Ops, RAG, HR Ops, Reporting and Support. Sourcedronahq.com/agents/templatesread 2026-10-02 |
|
| Triggers & Channel Coverage | Full |
|
Agents start on their own through triggers, and each can carry several at once, including a webhook, a scheduler, and 145+ DronaHQ platform triggers that fire on events in Gmail, Outlook, Slack, GitHub, Jira, Linear, Salesforce, HubSpot and Stripe, plus chat, WhatsApp, Instagram and Facebook Messenger for conversations. An email trigger is marked coming soon. Sourcedocs.dronahq.com/agents/triggers/introductionread 2026-10-02 |
|
| Model Flexibility & Routing | Full |
|
Each agent uses a model the customer chooses from OpenAI, Anthropic or Gemini, with DeepSeek, Cohere and xAI also listed, and providers and models can be mixed across agents. All models run on DronaHQ's built in key, and bring your own key is not offered. Sourcedronahq.com/agents-faqread 2026-10-02 |
|
| APIs, SDKs & MCP Extensibility | Full |
|
Agents connect to outside systems through APIs and webhooks, which DronaHQ calls the System interface, and developers get API keys, request logs and an agent API reference. The webhook trigger lets external systems start an agent with structured data. What the agent API covers is not described. Sourcedocs.dronahq.com/agents/getting-started/introductionread 2026-10-02 |
|
| Testing, Debugging & Optimization | Full |
|
In the evaluation harness, datasets hold test cases that specify context, the tool the agent must call and a reviewer for the final answer. Twelve reviewer types score outputs, including exact match, regex, JSON schema, Levenshtein, tone, factuality, closed QA and a custom LLM judge, and eval runs report average scores that can be compared side by side after a change. The agent pricing page still lists Evaluations as coming soon, while the docs describe the feature as live. Sourcedocs.dronahq.com/agents/getting-started/dataset-and-evalsread 2026-10-02 |
|
| Browser & Computer Use | Not documented |
|
The agent's built in tools are web search, time and calculator, and browser, desktop or remote computer control is not described. URL crawling feeds the knowledge base at build time and does not let the agent operate a browser. Sourcedocs.dronahq.com/agents/getting-started/introductionread 2026-10-02 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Pricing
$100 per user per month billed yearly ($140 monthly) on Starter, or pay as you go from $0.05 per conversation.
Per user plans with monthly AI credits and tool calls, or pay as you go per conversation plus model credits.
Included quota
Starter includes 5,000 AI credits and 10K tool calls a month, unlimited agents, 10 tools per agent, 30 day tracing and short term memory. Business includes 25,000 credits, 100K tool calls, unlimited tools, 90 day tracing and guardrails.
What is public
Pay as you go rates, plan card prices, credit and tool call allowances, add on prices and the credit rate are published, with per model credit rates in the docs. Enterprise is custom.
Billing mechanics
Plan cards bill per user per month, yearly or monthly, with monthly credit and tool call allowances. Pay as you go bills hosting per conversation and model use in credits at published per model rates.
Cost watchouts
Model cost is billed on top of hosting in credits, at a different rate per model and operation, including embeddings, vector search, file parsing and URL crawling. SSO, RBAC and SOC 2 sit in the Enterprise column of the pay as you go table, and zero data retention, 60 day retention and HIPAA are $1,000 a month add ons there.
Variable cost rationale
Every model call, embedding, vector search and crawl draws AI credits at per model rates, at $0.002 a credit, and pay as you go adds $0.05 per conversation for hosting. The bill grows with the workload, not with seats. Plan cards cap usage with monthly allowances, 5,000 credits and 10K tool calls on Starter and 25,000 credits and 100K tool calls on Business. Use beyond them is bought as top ups, 25,000 credits for $50 with rollover and 5,000 tool calls for $10 a month. Pay as you go has no monthly floor.
Additional watchouts
The $0.05 per conversation figure excludes the model, which bills separately in credits.
Overage / add-ons
Credits and tool calls top up as add ons without a plan change. A rollover top up of 25,000 AI credits costs $50, and 5,000 tool calls a month cost $10.
Sales call required
Mixed (some tiers require a call)
Free / trial
$5 free credit on pay as you go, and a free trial on the Starter and Business plans, length not published
Lowest paid plan
Starter at $100 per user per month billed yearly, or $140 monthly. Pay as you go has no monthly floor
Commercial notes
DronaHQ offers a self serve trial and pay as you go signup. Enterprise is an annual contract through sales. The agent platform is priced separately from the DronaHQ app builder.
Key ambiguities
The same page presents a pay as you go table and per user plan cards that do not reconcile, and the Business card marks Evaluations and A/B Testing as coming soon although the docs describe Evals as live.
Cancellation / refund
Plans bill yearly or monthly, and Enterprise is an annual contract through sales. No cancellation or refund terms are published.
Support SLA / resale
Enterprise includes priority 24x7 support and an SLA.
Missing data
No page says which pricing model a new self serve customer is placed on, or what the Enterprise floor is.
Related vendors
- Lindy — Lindy is an AI teammate that lives in Slack, answers from your tools…
- Activepieces — Open source, MIT licensed automation platform where every one of 700…
- Agenta — Agenta is an open source workspace where teams build AI agents by…
- AgentX — No code platform to build, evaluate, and deploy multi agent…
- AI Library — No code platform for building and running AI agents, with a…
- Aigensei — Enterprise agentic workflow platform that runs business processes…
Alternatives to DronaHQ
The closest documented capability profiles to DronaHQ among agent builders tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- Airtable11.0 / 14Fuller documented coverage on Workflow Orchestration
- Make11.0 / 14Fuller documented coverage on Workflow Orchestration
- Moveo.AI12.0 / 14Fuller documented coverage on Workflow Orchestration and Deployment & Data Residency
- SnapLogic12.0 / 14Fuller documented coverage on Workflow Orchestration and Deployment & Data Residency
- Haptik10.5 / 14Fuller documented coverage on Workflow Orchestration
- Mendix12.5 / 14Fuller documented coverage on Workflow Orchestration and Human Oversight & Guardrails
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded