Privacy Policy

Where we are the Data Fiduciary. We decide why and how we process the data needed to run our website and your account: your profile, sign-in and security records, billing records and your messages to us. For this data we are the Data Fiduciary.

Where we are a Data Processor. Businesses that use Dhanur AI Workforce put their own data into their workspaces: contacts, leads, conversations, documents, agent instructions and task records. For that data the business is the Data Fiduciary, and we act as its Data Processor. We process it only on the business’s instructions and under our Terms of Service.

If an agent run by one of our customers contacted you, please send your request to that business first. If you write to us instead, we will pass your request to the business and help it respond.

This policy does not cover third-party services that a customer connects, such as a messaging channel. Those services have their own privacy policies.

• Account and profile: your name, email address, workspace name, the use case and team size you tell us during sign-up, your role in each workspace, and invitations you send or receive (the invited email address, the role and who sent it).
• Google sign-in, if you use it: your name, email address, whether Google has verified that address, your profile picture and your Google account ID. We never receive your Google password.
• Sign-in and security data: hashed one-time sign-in codes; session records (a hash of the session token, the sign-in method, times, a hashed IP address and your browser details); a record of which version of our Terms and this policy you accepted, with a hashed IP address and browser details; and audit logs of security-relevant actions, such as creating a workspace or inviting a teammate.
• Workspace content: agent instructions, uploaded documents and knowledge, messages and conversations, details of the people your agents deal with, task records and outputs, and usage records such as which AI model ran a step and what it cost.
• Payment and billing records: your plan, invoices, payment status and transaction references from Razorpay, and the billing name, address and GSTIN you give us for invoices. Razorpay handles your card, bank and UPI details; we do not store them.
• Support and communications: what you send us by email, phone or in the product, and any feedback you give.
• Website: if you write to us or fill in a form on our website, the details you send. Our hosting and network providers also process technical data, such as IP addresses and browser type, to deliver pages and protect the site from attacks.

We use personal data to:

• create and run your account and workspaces, sign you in, and send sign-in codes and invitations;
• run your agents, show usage and costs, and enforce the approvals and spending limits you set;
• keep the Service secure, limit repeated sign-in attempts, and detect and investigate fraud or misuse;
• take payments, issue GST invoices and keep tax and accounting records;
• answer support requests and send service messages about security, billing and changes to our terms;
• fix problems and improve the Service using usage and cost records;
• comply with the law, respond to lawful requests and enforce our Terms.

We rely on your consent and on the legitimate uses allowed by section 7 of the DPDP Act, including where you have voluntarily given us data for a specific purpose and where we must comply with a law, court order or judgment. For business customers, we also process data to perform our contract with them.

We send promotional emails only if you agree to receive them, and you can opt out at any time. We do not sell personal data.

• We use one strictly necessary cookie, dh_session, to keep you signed in to the app. It holds a random token, and we store only a hash of that token. The cookie cannot be read by page scripts and is sent only over secure connections. It lasts until you sign out or for up to 30 days.
• The app saves interface preferences in your browser’s local storage, such as whether the sidebar is collapsed. This stays on your device.
• When you arrive through a link with an invite code, your browser keeps the code, and we add one to a count for that code and day. We store no cookie, IP address or other identifier with it.
• We use Google Analytics and Microsoft Clarity to understand how people use our website and app, so that we can find the parts that confuse them. Both set their own cookies and both process data outside India, under their own terms. We do not use advertising cookies, we have turned off Google’s advertising and cross-site features, and we do not sell anything to either of them.
• Three limits on that, which we hold ourselves to in code (src/lib/measurement.ts): neither tool runs on a share page or the website chat window, so your own customers are never recorded; inside the app every word is masked in a Clarity session replay, so we see where someone clicked and hesitated but not what they or their customers wrote; and page addresses are stripped of identifiers before they reach Google, so it receives “/tasks/[id]” rather than a link to your work.
• You can stop both at any time with your browser’s tracking protection or an ad blocker, and nothing in the Service depends on them.
• If you choose Google sign-in, Google may set its own cookies on its pages under its own policy.

• To perform a task, the Service sends the content that task needs, such as agent instructions, relevant documents and message text, to an AI model provider such as Anthropic. The provider processes it to return a result.
• We do not use your content to train AI models.
• We record which model ran each step and what it cost, so that we can bill accurately and show you what each task cost.
• Agents act on the instructions of the business that runs them. That business is responsible for decisions its agents make about people, and you can ask it for a human review.

We share personal data only with the service providers that help us run the Service, and only what each one needs:

• DigitalOcean: servers, databases and file storage. The servers that run our engine, and our databases, are in Bengaluru, India. Photos and PDFs that people send in chats are stored in DigitalOcean's New York region, in the United States.
• Vercel: hosting and delivery of our website and app.
• Cloudflare: DNS and network security.
• Postmark: transactional email, such as sign-in codes and invitations.
• Google: optional sign-in with Google.
• Razorpay: payment processing.
• Pipedream: connecting the apps a workspace chooses to link, and running the actions its agents are allowed to take in them. Only used by workspaces that connect apps.
• AI model providers, such as Anthropic: processing the content needed to perform a task.
• Brave: web search, for agents that a workspace lets search the web. Brave receives only the search words, keeps a record of them for up to 90 days for billing and troubleshooting, and does not link them to a person.
• Voyage AI: finding knowledge by meaning, by turning a workspace's knowledge and the questions asked of it into numbers that can be compared. We use Voyage's option that stops it keeping or training on what we send.
• fal.ai: making images, for agents a workspace lets create images. fal receives the description of the image and returns the image; for posters with printed words it runs Google's Gemini image model. We use fal's option that keeps no copy of either, and fal's terms say it does not use them to train its models.
• Pexels and Pixabay: free stock photos, for agents a workspace lets find photos for its posts. They receive only the few words the agent searches with, such as “team meeting in an office”, never anything about a customer, and we keep a copy of the photo the agent picks.
• Recall.ai: the meeting notetaker, for workspaces whose team sends one to a video meeting. Recall receives the meeting link and records the meeting, and turns what people say into a transcript, on servers in the United States. We delete the recording once we have the transcript, and it is deleted after 7 days at the latest. The notetaker says in the meeting that it is an AI recording it; the workspace that sends it is responsible for telling the people in its meetings.
• Apps a workspace connects directly (Google Calendar and Gmail, Slack, HubSpot, LinkedIn, and Instagram, which is run by Meta): only what the workspace uses them for, such as free and busy times and the bookings its agents make, emails it sends from its own Gmail, messages in the Slack channels our app is added to, leads it copies to HubSpot, the posts (and their images) a person approves for their own LinkedIn profile, and the photos and captions a person approves for their professional Instagram account. Meta fetches each approved photo from an unguessable link that stops working after an hour. From LinkedIn we receive only the connected person's name and member id, to show whose profile is connected and to post as them. From Instagram we receive only the account's id, username and type, to show which account is connected and to post to it, and the link to each post we publish. To delete a connected Instagram or LinkedIn account, see Delete your data.
• Messaging channels you connect, such as WhatsApp: delivering messages your agents send and receive.

We may also share personal data:

• with government authorities, courts or regulators when the law requires it, and where needed to protect people’s safety or prevent fraud;
• with a successor if our business is reorganised, merged or sold, on the condition that it keeps to this policy.

Inside a workspace, other members can see content and member details according to their roles. The workspace owner controls who has access.

Our servers and databases are hosted in India. Photos and PDFs that people send in chats are stored with DigitalOcean in New York, United States; they are private and are only served through our own service after an access check. Some of our other service providers, such as those for website delivery, email, sign-in, messaging and AI models, also process data outside India. Section 16 of the DPDP Act permits this, except to countries that the Central Government restricts, and we do not transfer personal data to those countries.

A workspace can connect a Google account so its agents can book appointments on its Google Calendar and send outreach emails from its Gmail. This section says exactly what we do with data from Google.

• What we access. For Google Calendar: the list of calendars the account owns, the times they are free or busy (never what the events are), and the events our agents create on those calendars. For Gmail: permission to send email from the account, which we use only to send the outreach emails and follow-ups the workspace sets up and approves. We cannot read, search or delete anything in the mailbox. We also receive the account's email address, to show which account is connected.
• How we use it. Only to provide the features the workspace switches on: offering free times, booking the appointment a customer picks, and sending the workspace's own emails. We do not use Google data for advertising, we do not sell it, and we do not use it to train or improve AI or machine learning models.
• How we store it. The access that Google grants us (the sign-in tokens) is encrypted in our database in Bengaluru, India. We keep the bookings and emails our agents make, as part of the workspace's tasks and campaigns, for as long as the workspace keeps them. We do not copy calendar events or the mailbox.
• Who else sees it. To write an email or offer times, the details needed for that one task are processed by our AI model provider, which does not train on them. Our people see Google data only when the workspace asks us for help, for security, or when the law requires it. We share it with no one else.
• How to stop it. Disconnect the account in the app under Integrations, and we delete the tokens straight away. If you remove Dhanur AI's access at myaccount.google.com/permissions instead, we delete them the next time an agent tries to use them.

Dhanur AI's use and transfer to any other app of information received from Google APIs will adhere to the Google API Services User Data Policy, including the Limited Use requirements.

• Sign-in codes are valid for 10 minutes and deleted about a week after they expire.
• Sessions end when you sign out or after 30 days, and their records are deleted soon after they expire.
• Account data is kept while your account is active and for up to 90 days after it is closed.
• Workspace content is kept as the business that owns the workspace instructs. When a workspace is closed, it can be exported for 30 days and is then deleted within 90 days.
• Invoices and payment records are kept for as long as tax and accounting laws require.
• Security and audit logs are kept for as long as needed to protect the Service and meet our legal obligations.

We may keep data for longer where a law, court order or ongoing dispute requires it.

• All traffic to our website, app and servers is encrypted in transit.
• In our application database, sign-in codes, session tokens and IP addresses are stored only as hashes. Sign-in codes are single-use and stop working after repeated wrong attempts.
• Access to data is controlled by role, both inside workspaces and within our team.
• Security-relevant actions are written to audit logs.
• Repeated sign-in requests from the same email address or network are rate-limited.

No system is completely secure. If a personal data breach occurs, we will inform the affected people and the Data Protection Board of India as the DPDP Act requires.

Under the DPDP Act you have the right to:

• access information about the personal data we process about you, how we process it and whom we have shared it with;
• correct, complete or update your personal data;
• erase personal data we no longer need, unless the law requires us to keep it;
• withdraw consent at any time, as easily as you gave it. Withdrawal does not affect processing that happened before it, but we may no longer be able to provide the Service to you;
• nominate another person to exercise your rights if you die or become unable to do so;
• have your grievances addressed by us.

To use these rights, write to outreach@prodigalai.com from your registered email address, or call +91 96548 13810. We may ask you to confirm your identity. We will respond within 30 days. For data inside a customer’s workspace, see clause 1.

Delete your data explains how to delete a connected Instagram or LinkedIn account yourself, and how to ask us to delete a workspace or your account.

If you are unhappy with how we handle your data, please contact our Grievance Officer first (details below). If our response does not resolve your complaint, you can then complain to the Data Protection Board of India.

The Service is not for anyone under 18, and we do not knowingly collect personal data from children. If we learn that we have, we will delete it.

If a customer’s workspace content includes personal data of children, the customer must obtain verifiable consent from a parent or lawful guardian as the DPDP Act requires, and must not track, monitor the behaviour of or target advertising at children.

We may update this policy. The date at the top shows when it last changed. Before a material change takes effect, we will tell you by email or in the product, and the app will ask you to accept the updated version.

For privacy questions, requests or complaints, write to outreach@prodigalai.com or call +91 96548 13810. Our Grievance Officer can be reached at the same email and phone number. Our company details are below.