Frequently Asked Questions
Have questions?
We have answers.
Everything you need to know about Rare Agents, your AI agents, and how it all works.
General
What is Rare Agents?
Rare Agents is a multi-agent AI platform that allows users to create, manage, and deploy their own AI agents through a structured dashboard. Each agent can act as a personal assistant, research assistant, or Solana-enabled execution assistant.
How is Rare Agents different from ChatGPT?
Rare Agents is not just a chat interface. It provides multi-agent management, tool control and permissions, on-chain Solana interaction, audit logs, workspace isolation, and approval-gated execution. You control what your agent can and cannot do.
What does "create your own agent" mean?
You can define the agent's name, its purpose, its identity instructions, which tools it can use, and whether it can interact with Solana. Each agent operates independently inside your account.
Is Rare Agents autonomous?
No. Rare Agents operates under user approval. Sensitive actions like sending SOL or writing data on-chain require your confirmation before execution.
Do I need technical experience to use Rare Agents?
No. The platform is designed for both beginners and advanced users. Developers can go deeper, but everyday users can operate through the dashboard without any coding knowledge.
Security & Privacy
Is my data private?
Yes. Each user's agents operate in isolated workspaces. No other user can access your data, memory, or configurations.
Does Rare Agents control my wallet?
No. Rare Agents never holds your private keys. Transactions are drafted and then signed by you using your connected wallet (Phantom, Solflare, etc.).
Can an agent send funds without my permission?
No. All financial transactions require user approval before signing. The platform is designed with an approval gate that cannot be bypassed.
Are actions logged?
Yes. Rare Agents provides an audit log that tracks agent messages, tool usage, on-chain actions, and system events.
Where are API keys stored?
Model API keys are stored securely and never shared between users. All sensitive credentials are encrypted at rest.
Solana Integration
What can agents do on Solana?
Currently agents can send memo transactions (write data on-chain) and draft SOL transfers (which require your signature). More Solana features may be added in future releases.
Which Solana network is supported?
Devnet is the default for safety. Mainnet-beta can be enabled by the user from the Solana tab inside each agent.
Why would I want to send a memo transaction?
Memo transactions allow you to permanently record small pieces of data on-chain. This can be useful for timestamping, business proof, event logging, and operational tracking.
Does Rare Agents pay gas fees?
No. You pay the transaction fee from your connected wallet. Gas fees on Solana are typically a fraction of a cent.
Agents & Tools
What tools can agents use?
Tools currently include: web search (real-time DuckDuckGo results), document analysis (PDF, Word, Excel, PowerPoint), Solana memo (write on-chain data), and SOL transfer preview (requires Pro or Elite plan). Future tools may include calendar, email, analytics, and third-party integrations.
Can I create multiple agents?
Yes. The number of agents you can create depends on your subscription plan — 1 on Free, 20 on Pro, and Unlimited on Elite.
Can agents share memory?
No. Each agent has fully isolated memory. This ensures your agents remain focused and their histories do not mix.
Can I delete an agent?
Yes. Agents can be permanently deleted from the Settings tab inside the agent detail page. This removes all associated messages, logs, and configuration.
Can I edit an agent's personality?
Yes. You can modify the agent's identity instructions at any time from the Memory tab. Changes take effect on the next message.
Can I stop an agent?
Yes. Agents can be paused or stopped from the Settings tab inside the agent. A stopped agent retains its history and configuration but will not respond to new messages.
Pricing & Plans
Is there a free plan?
Yes. The Free plan includes 3 agents, limited token usage, access to all AI models via the Smart Router, web search, document analysis, and Solana memo. No credit card is required.
What does the Pro plan include?
Pro includes 20 agents, Multi-Agent Collaboration, 10 million tokens per month, SOL transfers, persistent chat history, full audit logs, and priority performance for $14.99/month.
Do you charge per message?
Plans include token usage limits. Each message and response consumes tokens based on length. Additional usage may require a plan upgrade.
Can I cancel anytime?
Yes. Subscriptions can be managed or cancelled at any time through the Billing page in your dashboard.
Do you accept crypto payments?
Currently payments are processed via Stripe using a credit or debit card. Crypto-native billing may be added in the future.
Technical
What AI models does Rare Agents use?
Rare Agents has 11 live AI models: Qwen3.5 397B (largest, default), Kimi K2.6, Kimi K2.5, Qwen3 235B Thinking (deep reasoning), Qwen2.5 Coder 32B (code), Qwen3 32B (balanced), Qwen3.6 27B (fastest), DeepSeek V3.2 (stable), Google Gemma 4 31B, Claude claude-sonnet-4-5, and Claude 3.7 Sonnet (Premium). Open-source Chutes models run on Bittensor Network Subnet 64 via decentralised Trusted Execution Environments (TEE). A Smart Router automatically picks the best model for each query.
Can agents execute code?
Only controlled tool functions are allowed. Agents can use structured tools like web search and document parsing, but arbitrary shell execution is not permitted.
Is Rare Agents enterprise-ready?
Yes. The platform includes audit logging, permission gating, workspace isolation, and structured approval controls designed for professional use.
What is Rare Agents' long-term vision?
Rare Agents aims to become a structured AI agent infrastructure platform where users can deploy secure, permissioned, blockchain-aware AI agents at scale — giving individuals and teams full control over their autonomous AI systems.
AI Models
What is the Smart Model Router?
The Smart Model Router automatically analyses each of your messages and selects the most appropriate AI model. Simple factual questions go to Qwen3.6 27B for near-instant replies. Complex multi-step reasoning switches to Qwen3 235B Thinking. Code tasks route to Qwen2.5 Coder 32B. Everything else defaults to Qwen3.5 397B — the platform's largest and best general model. You can always override the router by pinning a specific model from your agent's Settings tab.
What is Qwen3.5 397B and why is it the default?
Qwen3.5 397B is Alibaba's latest 397-billion parameter model and the largest on the platform. It delivers the best overall quality for general tasks — research, writing, analysis, and complex instructions — making it the Smart Router's default choice for standard queries. Its scale means it handles nuanced instructions better than smaller alternatives.
What are Kimi K2.6 and K2.5?
Kimi K2.6 and K2.5 are Moonshot AI's flagship models, known for strong instruction-following and reasoning across multi-turn conversations. K2.6 is the larger of the two and serves as the primary intelligent fallback after Qwen3.5 397B in the routing chain. K2.5 is slightly lighter and faster.
What is Qwen3 235B Thinking best used for?
Qwen3 235B Thinking is a 235-billion parameter MoE model with an internal reasoning (thinking) mode. It is auto-selected by the Smart Router for complex queries — deep research synthesis, multi-step logical problems, and tasks where thoroughness matters more than speed. It runs inside a Trusted Execution Environment (TEE) on Bittensor Subnet 64.
What is Qwen2.5 Coder 32B?
Qwen2.5 Coder 32B is purpose-built for software engineering tasks — code generation, debugging, refactoring, and technical explanations. The Smart Router automatically routes any message detected as a coding task to this model. You can also pin it manually for a dedicated coding session.
What does Bittensor Subnet 64 mean for my privacy?
All Chutes AI models on Rare Agents (Qwen, Kimi, DeepSeek, Gemma) run on Bittensor Network Subnet 64 — a decentralised, privacy-preserving compute network. Inference happens inside Trusted Execution Environments (TEEs), which provide hardware-level proof that your prompts and the model weights are not visible to the infrastructure operator. This is a stronger privacy guarantee than traditional centralised cloud inference.
Can I switch AI models mid-conversation?
Yes. You can change an agent's active model at any time from the agent's Settings tab. The change takes effect on the next message. Conversation history is preserved — the new model will see the full prior context. You can also override the Smart Router for a session by pinning a specific model.
Why was DeepSeek R1 removed?
DeepSeek R1 was removed because its compute cost made it unsustainable at subscriber pricing. It has been superseded by more capable models — Qwen3 235B Thinking now exceeds R1's reasoning quality at a fraction of the inference cost.
Browser Automation
How does browser automation work on Rare Agents?
Rare Agents browser automation uses a Chrome extension that runs on your computer and connects to your agent via a secure WebSocket. When you give your agent a browser task — like 'fill out this form', 'find the price on this page', or 'click the submit button' — the agent sends step-by-step instructions to the extension, which carries them out in a real Chrome tab. No remote browser infrastructure is required.
Do I need to install anything for browser automation?
Yes — a one-click Chrome extension install is required. Once installed, it automatically connects to your agent's session. The extension is lightweight and only activates when you start a browser automation task.
What kinds of browser tasks can an agent perform?
Agents can navigate to URLs, click buttons and links, fill in text fields and forms, read the content of any webpage, take screenshots, scroll, select dropdown options, and extract structured data from pages. This makes them capable of research, web scraping, automated form submission, data entry, and website monitoring.
Is browser automation available on the free plan?
Yes. Browser automation is available on all plans including the free tier. You can start using it immediately after installing the Chrome extension — no upgrade required.
Can the agent browse any website?
Agents can access any publicly reachable website. Sites that require login can also be automated if you are already logged in on that Chrome session. Rare Agents does not bypass security measures, CAPTCHAs, or terms of service of third-party websites — users are responsible for complying with the rules of sites they automate.
Data Vault
What is the Personal Data Vault?
The Personal Data Vault is a private, encrypted storage space built into your Rare Agents account. Any text, note, credential, research result, or AI-generated content can be saved there. Every item is encrypted with AES-256-GCM before it reaches our servers — only you hold the decryption capability through your account.
What encryption standard does the vault use?
The vault uses AES-256-GCM — the same encryption standard used by banks, the US government, and financial institutions worldwide. The encryption key for each user is derived using scrypt, a memory-hard key derivation function, making brute-force attacks extremely expensive. Not even the Rare Agents team can read your vault contents.
What does Solana blockchain proof-of-existence mean for my vault items?
When you save an item to the vault, a SHA-256 cryptographic hash (fingerprint) of that item's content is recorded as a memo transaction on the Solana blockchain. This creates a permanent, timestamped, publicly verifiable record that the content existed at that moment — without revealing the content itself. If anyone later questions whether you held that information on a certain date, the blockchain proves it.
How many items can I store in the vault?
The Free plan supports up to 10 vault items. The Pro plan supports up to 100 items. The Elite plan has no hard limit. Each item can store a substantial amount of text, so even 10 items are enough for critical credentials, research notes, or important documents.
What is Smart Data Monetization?
Smart Data Monetization is an AI-powered feature that automatically analyzes your agent's chat responses and identifies content that is likely to have commercial value — such as research insights, market analysis, or unique findings. When it detects valuable content, it suggests saving it to your vault and optionally listing it on the Data Marketplace for others to license.
Can I export vault items?
Yes. Vault items can be viewed and copied from the vault interface. The decrypted content is displayed only when you are logged in. You can copy the content to any external system at any time.
Messaging Channels
Can I connect my AI agent to WhatsApp?
Yes. Rare Agents supports WhatsApp through the open Baileys protocol (no WhatsApp Business API or paid account required). You connect by scanning a QR code from the Messaging tab inside your agent. Once connected, your agent automatically reads incoming WhatsApp messages and responds using your chosen AI model and personality.
Can I connect my AI agent to Telegram?
Yes. To connect Telegram, create a bot using @BotFather on Telegram, copy the token it gives you, and paste it into the Messaging tab of your agent. Your agent will begin responding to all messages sent to that Telegram bot in real-time.
Does the agent respond automatically or do I have to approve each reply?
Messaging channel responses are automatic by default — the agent reads the incoming message and immediately replies using your AI model. This makes your agent a true 24/7 responder on WhatsApp and Telegram. You can review all incoming and outgoing messages in the agent's chat history.
Is my WhatsApp session data secure?
Yes. WhatsApp session credentials (the authentication state that keeps you connected without re-scanning the QR code) are encrypted before being stored. The session is scoped to your agent only and is never shared with other users.
Can I have both WhatsApp and Telegram connected to the same agent?
Yes. Each agent supports one WhatsApp connection and one Telegram bot connection simultaneously. When a message comes in from either channel, the agent responds through the same channel it was received on.
Data Marketplace
What is the Data Marketplace?
The Data Marketplace is a built-in platform within Rare Agents where users can list vault items for sale or licensing. If your AI agent has collected research, analysis, reports, or other valuable content that you have stored in your vault, you can make it available for others to purchase or subscribe to.
How do I create a marketplace listing?
From the Data Marketplace tab in your dashboard, click 'New Listing', select a vault item, set a title, description, category, price, and choose a license type (single purchase, subscription, or open license). Once published, other Rare Agents users can discover and buy it.
What types of data can I list on the marketplace?
Any vault item can be listed — research reports, AI conversation summaries, curated datasets, written content, market analysis, tutorials, templates, or any other text-based information. The content must be original or data you own the rights to distribute.
Who can buy marketplace listings?
Currently, marketplace listings are visible and purchasable by other Rare Agents users. The marketplace is a closed ecosystem within the platform for trusted data exchange.
Can I generate AI images or videos with my agent?
Yes. Just type a natural-language request in any agent's Chat tab — "generate an image of…" or "create a video of…" — and the result appears inline in the conversation. Images use the FLUX.1 model (Hugging Face) and videos use MiniMax Video-01 (Replicate). You can also use the 🎨 button for manual prompt control. The HF_API_TOKEN and REPLICATE_API_TOKEN secrets power these integrations.
Integrations
What OAuth connectors does Rare Agents support?
Rare Agents supports OAuth-based connections to four major services: Google (Gmail, Drive, Calendar data sync), Slack (channel messaging and workspace data), Notion (pages and database sync), and GitHub (repository and activity data). Each connector lets your agent read and optionally write data to these platforms on your behalf.
How do I connect Google to my agent?
From the OAuth Connectors tab in your dashboard, click Connect next to Google and authorize through the standard Google OAuth flow. Once connected, your agent can read your Gmail summaries, Google Drive file metadata, and calendar events to inform its responses.
Does connecting Slack give the agent full access to my workspace?
No. The connector requests only the specific OAuth scopes needed for the features you enable. For Slack, this typically covers reading messages in designated channels and posting on your behalf. You can review and revoke the connection at any time from the Integrations tab.
Does Rare Agents support two-factor authentication?
Yes. Rare Agents supports TOTP (Time-based One-Time Password) two-factor authentication, compatible with apps like Google Authenticator, Authy, and 1Password. When you enable 2FA, you also receive 8 one-time backup codes to store safely in case you lose access to your authenticator app. 2FA can be enabled from Account Settings.
Agent Security Shield
What is the Agent Security Shield?
The Agent Security Shield is a proprietary, server-side prompt security layer built into every agent on Rare Agents. It automatically scans every incoming message — whether from the dashboard, WhatsApp, Telegram, or any API — and scores it for malicious intent before the AI ever processes it. No other mainstream AI platform does this at the agent level.
What is the 20-point injection detection system?
Rare Agents' 20-point detection system is a multi-pattern, weighted scoring engine that evaluates every message across 20+ distinct attack signatures. Each pattern targets a specific technique attackers use to subvert AI agents. The system assigns a weighted score per match — lighter patterns score lower, high-confidence attack signals score higher — and the combined total determines whether a message is Low, Medium, or High severity.
What six attack categories does the shield cover?
The shield covers: (1) Instruction Override — attempts to overwrite the agent's system prompt with phrases like 'ignore all previous instructions'; (2) Persona Hijacking — telling the agent to 'pretend you have no restrictions' or 'act as DAN'; (3) System Prompt Extraction — asking the agent to reveal or repeat its hidden instructions; (4) Role Delimiter Injection — using special tokens like <system> or [INST] to escape the conversation context; (5) Data Exfiltration — prompts engineered to make the agent leak user data, vault contents, or API keys; (6) Jailbreak Social Engineering — layered manipulation attempts that try to bypass guidelines through fictional framing or hypothetical scenarios.
What is the difference between Standard and Strict mode?
In Standard mode (the default), the shield logs all detected threats and flags the message in your audit log, but allows the conversation to continue. This gives you full visibility without interrupting users. In Strict mode, any message scoring at Medium or High severity is actively blocked before it reaches the AI — the agent returns a rejection notice instead of processing the request. Strict mode is recommended for agents exposed to public or untrusted audiences, like those connected to WhatsApp or Telegram.
Can the shield stop jailbreak attempts?
Yes. The shield is specifically trained on known jailbreak patterns — including DAN prompts, role-play bypass attempts, token injection, and multi-step social engineering chains. When Strict mode is active, these attempts are blocked outright. In Standard mode they are logged and flagged so you can review them. No filter is 100% perfect, but the 20-point system catches the vast majority of documented and common jailbreak techniques.
What happens when a threat is detected?
In Standard mode: the message is processed normally, but a threat record is written to your audit log with the detected patterns, severity score, and timestamp. You can review all threats from the Activity page. In Strict mode: the message is rejected and the agent replies with a security notice. The event is still logged so you have a full record. High-severity blocks are highlighted prominently in your activity feed.
Is the Agent Security Shield enabled by default?
Yes. Every agent starts with the shield in Standard mode automatically — no setup required. You can upgrade any agent to Strict mode from the Security tab inside the agent detail page. You can also disable the shield entirely for agents where you want unrestricted input, though this is not recommended for publicly-facing agents.
Social Media Agents
What is the Social Media Agent system?
The Social Media Agent system is a built-in social media management layer inside Rare Agents. It allows your AI agents to learn your brand voice, generate post candidates, manage an approval queue, and automatically improve over time based on what content you accept or reject. It turns any agent into a dedicated, AI-powered social media manager.
Which social media platforms can I connect?
The Social Media system supports X (Twitter), YouTube, Facebook, and Instagram via OAuth-based account connections. Once connected, your agent can generate platform-specific content, schedule posts, and track performance data for each channel — all from a single unified hub inside your agent.
What is Brand Voice Training and how does it work?
Brand Voice Training is the second step in the Social Media setup flow. You teach the agent your brand's tone (professional, casual, witty, authoritative), core topics, banned phrases, and example posts you love. The agent uses this training data as its creative brief whenever it generates content — so every post candidate matches your voice, not a generic AI tone.
What is the Command Center?
The Command Center is the main operational tab of the Social Media system. It shows your approval queue — AI-generated post candidates waiting for your review — along with a live event feed of social activity. Each candidate includes a risk classification (Safe, Caution, Review) so you can quickly identify which posts need more attention before publishing.
What is the approval queue and why does it matter?
The approval queue ensures you stay in control. Your agent generates post candidates automatically, but nothing is published until you approve it. You can approve a post as-is, edit it before approving, or reject it with optional feedback. Rejected posts — and your edits — are fed back into the Learning Dashboard so the agent improves its suggestions over time.
How does the AI generate post candidates?
The agent uses your Brand Voice Training data, connected platform context, and a 4-model fallback chain to generate post candidates. It considers your tone, banned phrases, example content, and platform-specific best practices (character limits, hashtag conventions, optimal formats). Each candidate is scored for risk before being placed in the approval queue.
What is the Learning Dashboard?
The Learning Dashboard tracks your agent's content performance over time. It shows your acceptance rate (how often you approve AI suggestions), your edit rate (how often you modify posts before approving), per-platform statistics, and improvement trends. The agent reads these signals as a feedback loop — the more you use the system, the more precisely it learns what content you want.
vs. Competitors
How is Rare Agents fundamentally different from ChatGPT?
ChatGPT is a single conversational interface — one session, one assistant, no persistence, no agents, no tool control. Rare Agents gives you a multi-agent platform where every agent has its own name, personality, memory, isolated tool permissions, audit log, and Solana wallet. You can run up to 20 (Pro) or unlimited (Elite) specialized agents simultaneously, connect them to WhatsApp and Telegram, store and monetize their outputs, and protect them with a 20-point security shield. ChatGPT has none of these capabilities.
How does Rare Agents compare to Claude?
Claude (Anthropic) is a conversational AI focused on safety and helpfulness in single-session interactions. Rare Agents is an agent management platform — not just a chat interface. While Claude excels at in-depth reasoning within a conversation, Rare Agents gives you persistent agents with memory across sessions, automated social media management, blockchain-anchored data storage, a marketplace to monetize AI outputs, and browser automation. Claude has no agent lifecycle management, no data vault, and no monetization layer.
How does Rare Agents compare to Perplexity AI?
Perplexity is primarily a web-search-powered answer engine — it retrieves and summarizes live web results. Rare Agents does include real-time web search as one of its agent tools, but it goes far beyond search. Your agents retain memory, run on WhatsApp and Telegram, encrypt and anchor data to the blockchain, generate and monetize content, and are protected by a multi-layer security shield. Perplexity has no agent management, no data ownership layer, and no monetization capability.
How does Rare Agents compare to Google Gemini?
Google Gemini is Google's AI assistant, tightly integrated into the Google ecosystem. Rare Agents is platform-agnostic — it connects to Google, Slack, Notion, GitHub, WhatsApp, Telegram, and more, while keeping you in control of your data and outputs. More importantly, Rare Agents lets you own, encrypt, blockchain-anchor, and sell the AI-generated data your agents produce. Google monetizes your data. Rare Agents helps you monetize it yourself.
Can I own and sell the AI outputs my agents produce?
Yes — and this is one of Rare Agents' most distinctive capabilities. Any content, research, analysis, or insights your agents generate can be saved to your encrypted Personal Data Vault, blockchain-anchored on Solana to prove authenticity and timestamp, and listed on the Data Marketplace for other users to license. No other mainstream AI platform — ChatGPT, Claude, Gemini, or Perplexity — gives users any form of data ownership or monetization.
Why should I use Rare Agents instead of just using multiple AI tools?
With generic AI tools, you juggle multiple tabs, lose conversation history, manually copy outputs between systems, and have no security layer, no data ownership, and no monetization path. Rare Agents unifies everything: multiple specialized agents in one dashboard, persistent memory, built-in security, blockchain-verified storage, social media automation, browser control, and a data marketplace — all managed from one account. It replaces a fragmented AI workflow with a structured, auditable, monetizable system.
Data & Blockchain
How does the blockchain proof-of-existence actually work?
When you save a vault item, the platform computes a SHA-256 cryptographic hash of the content — a unique 64-character fingerprint that changes if even one character of the content changes. That hash is then written into a Solana Memo Program transaction and committed to the blockchain. The transaction is permanent, timestamped, and publicly verifiable on Solscan. This proves the exact content existed at that exact time — without ever revealing the content itself.
Can I independently verify a vault item on the blockchain?
Yes. Every vault item with a blockchain record shows a Solana transaction signature in the vault interface. You can paste that signature into any Solana block explorer (like Solscan or Solana Explorer) and see the exact memo, timestamp, and block confirmation. To verify integrity: take the original content, compute its SHA-256 hash locally, and compare it to the memo on-chain. If they match, the content is provably unchanged since it was anchored.
Can the Rare Agents team read my vault contents?
Vault content is encrypted server-side with AES-256-GCM using account-specific keys derived from a server-held master secret. This is not zero-knowledge or end-to-end encryption: the service can decrypt content for authorized features. Information used in AI tasks may be processed by the selected model provider. Review the privacy policy before storing sensitive information.
What happens to my vault data if I cancel my subscription?
Your vault data and blockchain records are not deleted when you downgrade or cancel. The blockchain records are permanent by nature — no one can remove them. Vault access and encrypted data storage are retained. If you downgrade below your current item count limit, you will not be able to add new items until you are within the new plan's limit, but existing items remain accessible.
Why is Solana used for blockchain anchoring instead of Ethereum?
Solana was chosen for three reasons: (1) Transaction fees are a fraction of a cent — anchoring data costs almost nothing compared to Ethereum's gas fees; (2) Finality is near-instant (under a second) so your proof is confirmed immediately; (3) Memo transactions are a standard, well-supported Solana primitive with native support in block explorers. Ethereum would make the same feature 100x more expensive and significantly slower.
Does blockchain anchoring change or reveal my data?
No. The blockchain only stores a hash — a mathematical fingerprint. The hash is one-directional: you cannot derive the original content from it. The full content never leaves your encrypted vault. The blockchain record proves the content existed and hasn't been altered, without exposing a single word of what the content actually says.
Selling Your Data
Who can buy data on the Rare Agents marketplace?
Currently the marketplace is available to all Rare Agents users. Listings are discoverable within the platform by other registered users. As the marketplace grows, public discovery features will expand. Pro and Elite plan holders can create unlimited listings; Free plan users have a listing limit.
How should I price my data listing?
Pricing depends on the depth and uniqueness of your content. Niche research reports, curated datasets, domain-specific analysis, and rare market insights command higher prices. Generic summaries or widely available information tend to sell less. Start by reviewing similar listings on the marketplace for benchmarks. You can adjust pricing after publishing — it applies to new purchases immediately.
What license types are available when I list data?
When creating a listing you choose from three license types: (1) Single Purchase — the buyer gets one-time access to the content; (2) Subscription — buyers pay a recurring fee for ongoing access, useful for content you update regularly; (3) Open License — the content is freely available but your authorship and blockchain timestamp are preserved. You can also set whether buyers can redistribute the content or use it for commercial purposes.
What kind of data tends to perform best on the marketplace?
The highest-value listings tend to be: detailed research reports on specific industries or topics, curated datasets with original analysis, AI-generated competitive intelligence, niche how-to guides with structured steps, and unique market or price trend summaries. Content that is hard to find elsewhere, clearly structured, and backed by a blockchain authenticity record performs best. The Smart Data Monetization feature automatically flags your AI outputs that meet these criteria.
Can I take a listing down after publishing it?
Yes. You can unpublish or delete a listing at any time from the Data Marketplace tab. Unpublishing hides the listing from discovery but preserves any existing purchases — buyers who already purchased retain their access. Deleting removes the listing entirely. The underlying vault item and its blockchain record are not affected by listing changes.
What is Smart Data Monetization?
Smart Data Monetization is an AI-powered heuristic that runs automatically after every agent response. It analyzes the content of the response and scores it for commercial value — looking for signals like unique insights, structured analysis, proprietary research, or original findings. When it detects high-value content, a prompt appears suggesting you save it to your vault and optionally list it on the marketplace. It removes the manual work of deciding what's worth monetizing.
Still have questions?
Reach out to our team or launch the dashboard and start exploring.