canonical: https://jentic.com/apis/aidbase/aidbase # Aidbase API Jentic publishes the only available OpenAPI specification for Aidbase API, keeping it validated and agent-ready. Aidbase is an AI-powered customer-support platform that helps businesses enhance self-serve and automate support. This spec covers a richer surface than the public Aidbase v1 - knowledge management with website, video, document, and FAQ ingestion, chatbots, ticket forms, tickets, and email inboxes. Authentication uses HTTP Bearer tokens. The base URL is https://api.aidbase.ai/v1. ## For AI agents Ingest knowledge from websites, videos, documents, and FAQs, manage chatbots and ticket forms, and triage tickets and email inboxes through a Bearer-authenticated REST surface. ## Scope Does not handle outbound marketing, voice channels, or general CRM contact management - use for AI knowledge ingestion, support chatbots, and ticketing only. ## Capabilities - Ingest knowledge from a website URL, video, uploaded document, or FAQ entry through the /knowledge endpoints - Finalize a document-based knowledge item with `/knowledge/{id}/finalize` once upload completes - Inspect knowledge sub-pages and FAQ items derived from an ingested source - Manage chatbots, their bindings, and their conversations across the chatbot endpoints - Create ticket forms with custom fields and accept submissions into the /tickets resource - Connect email inboxes so inbound mail enters the same ticketing workflow ## Use cases ### Multi-Source Knowledge Ingestion Build a knowledge base that draws from a help-centre website, a set of how-to videos, a few PDF manuals, and a hand-written FAQ. Use POST `/knowledge/website`, `/knowledge/video`, `/knowledge/document`, and `/knowledge/faq` with `/knowledge/{id}/finalize` to wrap up document uploads. The result is a unified knowledge surface a chatbot can answer from. Example prompt: POST `/knowledge/website` with the help-centre URL, then POST `/knowledge/document` for each PDF and call `/knowledge/{id}/finalize` once each upload completes ### Self-Serve Support Chatbot Stand up a self-serve chatbot that answers from the ingested knowledge. The chatbot endpoints let you create the bot, bind it to knowledge, and send chat messages. Combined with sub-pages and FAQ items derived from sources, the bot can cite the original article that produced its answer. Example prompt: Create a chatbot bound to the knowledge items ingested above, then send the customer's question to its chat endpoint and capture the cited sources ### Custom Ticket Forms with Email Intake Combine ticket forms and email inboxes so customers can submit structured tickets via a form or simply reply to a support email. Tickets are listed and updated through the /tickets endpoints, and email inboxes route inbound mail into the same pipeline. Example prompt: Create a ticket form, register an email inbox, and configure the inbox to route inbound mail into the form to create tickets ### Operational Knowledge Audit Periodically audit which knowledge items power the chatbot by listing /knowledge, then drilling into `/knowledge/{id}/sub-pages` and `/knowledge/{id}/faq-items.` Items that no longer match the live product can be deleted via DELETE `/knowledge/{id}` so the chatbot stops citing stale content. Example prompt: GET /knowledge, then for each item GET `/knowledge/{id}/sub-pages` and decide whether to keep or DELETE the item ### Agent-Driven Support via Jentic An AI agent automates support: ingest a new help article, retrain knowledge, and triage incoming tickets. Through Jentic the agent finds the right Aidbase operation by intent, the Bearer token is supplied from the vault, and the agent never sees the raw token. Example prompt: Use Jentic search 'ingest a website into a knowledge base', execute `/knowledge/website`, then list /tickets and post a draft reply on each ## Key endpoints | Method | Path | Description | | --- | --- | --- | | GET | `/knowledge` | List all knowledge items | | POST | `/knowledge/website` | Ingest a website as a knowledge source | | POST | `/knowledge/document` | Create a document knowledge item | | POST | `/knowledge/{id}/finalize` | Finalize a document knowledge upload | | POST | `/knowledge/faq` | Create an FAQ knowledge item | | GET | `/knowledge/{id}/sub-pages` | List sub-pages derived from a knowledge source | | GET | `/knowledge/{id}/faq-items` | List FAQ items derived from a knowledge source | ## Key resources - **Knowledge** — Ingestion endpoints for website, video, document, and FAQ sources, plus item-level inspection - **Chatbots** — Chatbot management and chat interactions backed by ingested knowledge - **Ticket Forms** — Custom-field forms that produce tickets - **Tickets** — Customer-support tickets with listing, update, and lifecycle management - **Email Inboxes** — Inbound email routing into the ticket pipeline ## AI readiness This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use. - **Score:** 67 / 100 - **Maturity:** AI-Aware - **Dimensions:** - Foundational Compliance: 89 / 100 - Developer Experience & Jentic Compatibility: 63 / 100 - AI-Readiness & Agent Experience: 47 / 100 - Agent Usability: 94 / 100 - Security: 60 / 100 - AI Discoverability: 100 / 100 - **View full report:** https://jentic.com/apis/aidbase/aidbase/scorecard - **How the score is calculated:** https://docs.jentic.com/reference/api-readiness-framework/overview/ - **More about the dimensions:** https://docs.jentic.com/reference/api-readiness-framework/specification/#dimensional-model-overview ### Score it yourself Every API in the directory is allowlisted, so you can re-score it with no key required. - **Score your own API:** https://jentic.com/scorecard.md - **Scoring CLI agent skill:** https://github.com/jentic/jentic-api-scorecard/blob/main/skills/jentic-api-scorecard/SKILL.md ```sh npx @jentic/api-scorecard-cli score ``` ## Why Jentic - **Setup:** Wiring Aidbase by hand means handling its bearer token, sequencing ingestion calls like `/knowledge/website` then `/knowledge/{id}/finalize`, and writing your own retry logic. Through Jentic you install once, import Aidbase from the API Directory, store the token once, and your agent calls it. - **Permission scoping:** Aidbase puts the knowledge source id in the URL path (`/knowledge/{id}/finalize`, `/knowledge/{id}/sub-pages`), so a rule can pin your agent to one knowledge source. You choose the operations it may call, so ingesting a website or document is only included if you add it. - **Credential handling:** Your Aidbase bearer token is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context. - **Discovery method:** Agents search Jentic by intent such as 'ingest a website into a knowledge base', and Jentic returns the Aidbase `/knowledge/website` operation with its source URL schema so the agent calls the right endpoint without browsing the reference docs. ## Related APIs - **Aidbase API** — Earlier published Aidbase surface with a different base path and apiKey scheme - **Intercom API** — Customer messaging and support platform - **Zendesk API** — Mature help-desk with tickets, users, and macros ## FAQ ### Why is there no official OpenAPI spec for Aidbase API? Aidbase does not publish an OpenAPI specification. Jentic generates and maintains this spec so that AI agents and developers can call the Aidbase API via structured tooling. It is validated against the live API and kept up to date. Get started with Jentic One, the self-hosted execution layer. ### What authentication does the Aidbase API use? HTTP Bearer authentication. The Bearer token goes in the Authorization header. Through Jentic the token is held in the vault and never enters the agent's context. ### Can I ingest a website into a knowledge base? Yes. POST `/knowledge/website` with the source URL. Aidbase crawls the page tree, and you can later inspect the resulting sub-pages via GET `/knowledge/{id}/sub-pages.` ### How do I upload a PDF as a knowledge document? POST `/knowledge/document` to create the document, upload the file, then call POST `/knowledge/{id}/finalize` to mark ingestion complete and make the document available to chatbots. ### What are the rate limits for the Aidbase API? The spec does not declare explicit rate limits. Use list endpoints with reasonable page sizes when iterating over knowledge or tickets, and respect HTTP error codes returned by the server. ### How do I drive Aidbase from Jentic? Run pip install jentic, search 'ingest a website into a knowledge base', execute `/knowledge/website`, and chain in `/knowledge/{id}/finalize` for any document uploads. Run it through Jentic One, the self-hosted execution layer. ### Can I limit what my agent is allowed to do with the Aidbase API? Yes. Because you run Jentic One yourself, your own rules decide which Aidbase operations and credentials the agent may use, so you can allow read-only calls like GET /knowledge and GET `/knowledge/{id}/sub-pages` while withholding ingestion operations such as POST `/knowledge/website` or POST `/knowledge/document.` Aidbase also carries the knowledge source id in the URL path (`/knowledge/{id}/finalize`, `/knowledge/{id}/sub-pages`), which lets a rule pin the agent to a single knowledge source. The agent can only call the operations you have explicitly permitted, and the bearer token stays with your own instance rather than the agent's context.