> For the complete documentation index, see [llms.txt](https://docs.useagentdock.xyz/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.useagentdock.xyz/resources/faq.md). # FAQ *** ## General **What is AgentDock?** AgentDock is a decentralized knowledge marketplace built on Solana. Knowledge creators publish proprietary datasets and domain-specific corpora as knowledge bases, and AI developers and agents purchase access to query them via a semantic retrieval API. Every transaction is settled on-chain. **Who is AgentDock for?** AgentDock is for two groups. Knowledge creators: researchers, data providers, institutions, and domain experts who want to monetize specialized knowledge. AI developers and agent operators: teams building vertical AI applications and agents that need domain knowledge beyond what public datasets provide. **Is AgentDock open source?** The AgentDock on-chain program will be open source. The platform and API are proprietary. The JavaScript SDK is open source. *** ## Pricing and payments **What currency is used for payments?** SOL during the beta. The INTL utility token launches at general availability and will offer discounted rates. **Where do my earnings go as a creator?** Directly to your connected Solana wallet. Earnings accumulate in an on-chain escrow account managed by the AgentDock program and can be withdrawn at any time. **Are there any platform fees?** Yes. AgentDock charges a protocol fee on each transaction. The current rate is shown in the creator dashboard. There are no listing fees during the beta. **Can I get a refund if a knowledge base does not meet my expectations?** Not currently. All purchases are final once confirmed on-chain. Review the listing preview and buyer reviews before purchasing. *** ## Technical **What retrieval latency should I expect?** The p95 target for retrieval calls is under 200ms. Actual latency depends on knowledge base size, query complexity, and network conditions. **Can I use AgentDock without the JavaScript SDK?** Yes. The REST API is fully documented and does not require the SDK. See the [API Reference](/api-reference/overview.md). **Can I query multiple knowledge bases in the same request?** Not in a single API call. Make parallel requests to each listing and merge results on the client side. See [Integrating Knowledge Bases](/for-buyers-and-developers/integrating.md) for an example. **What embedding models are used?** The default model is `text-embedding-3-large`. Creators can specify an alternative supported model during upload. If you upload pre-built embeddings, specify the model used so retrieval queries are embedded with the same model. **Is the retrieval API compatible with OpenAI function calling?** Yes. The API schema follows OpenAI tool/function calling conventions. You can register the retrieval endpoint as a tool in any compatible framework without an adapter. See [Integrating Knowledge Bases](/for-buyers-and-developers/integrating.md). *** ## Creators **Can I update a knowledge base after publishing it?** Yes. Content can be updated at any time. For one-time unlock listings, updates only apply to new purchases. For subscription listings, all active subscribers see the updated content immediately. See [Managing Listings](/for-knowledge-creators/managing-listings.md). **Can I set different prices for different buyers?** Not currently. Pricing is set at the listing level and applies to all buyers equally. **What happens if I delete a listing with active subscribers?** You cannot delete a listing with active subscribers; the unpublish action invalidates all credentials including those of active subscribers. Use the pause option to stop new purchases while keeping existing access intact. *** ## Protocol **Is AgentDock permissionless?** During the beta, creator applications go through a review process. Open beta and general availability will relax this to automated quality checks. The on-chain program itself is permissionless: anyone can call its instructions directly. **Can I verify my access credential without using the AgentDock API?** Yes. Access credentials are on-chain PDAs. You can derive the PDA address for any wallet and listing, read its state via any Solana RPC, and verify it yourself. See [Access Credentials](/protocol/credentials.md). **What happens to my data if AgentDock shuts down?** Listing metadata is stored on Arweave or IPFS, which are permanent and content-addressed. The on-chain program state is immutable. Your ability to use the retrieval API depends on AgentDock operating the API server. We recommend keeping local copies of any knowledge base content you upload. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.useagentdock.xyz/resources/faq.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.