canonical: https://jentic.com/apis/conduit.ai/conduit

# Conduit API

The Conduit API runs AI chat agents that handle customer conversations across tickets, contacts, and knowledge bases. It exposes conversations and their messages, helpdesk and support tickets with resolve, close, and reopen actions, contact records with proactive messaging, and a searchable knowledge base of nodes. Agents can be created, run, and simulated, and the API reports automation and resolution rates so a team can measure how much support work the agents absorb.

## For AI agents

Run AI support agents over customer conversations, manage tickets and contacts, search the knowledge base, and read automation and resolution metrics. Authenticated with a bearer token.

## Scope

Does not handle payment processing, outbound marketing campaigns, or phone-network provisioning. Use for AI-assisted customer-support conversations, tickets, and contacts only.

## Capabilities

- Run and simulate AI chat agents against a conversation and poll their runs
- Resolve, close, and reopen support tickets and post messages to them
- Send a proactive message to a contact and read their tickets, calls, and appointments
- Search the knowledge base and create or update knowledge nodes
- Read automation-rate and resolution-rate insights for the workspace
- List and search conversations and read their message history

## Use cases

### Automated Ticket Resolution

A support agent reads an open ticket, searches the Conduit knowledge base for the relevant answer, posts a reply, and resolves the ticket when the issue is handled. Conduit exposes the ticket's message history and resolve, close, and reopen actions, so the agent can work a ticket end to end and hand off only the cases that need a human.

Example prompt: Search the knowledge base with POST `/v1/kb/search`, post the answer with POST `/v1/tickets/{id}/messages`, then call POST `/v1/tickets/{id}/resolve`

### Proactive Contact Outreach

An agent identifies a contact that needs follow-up, reads their recent tickets and conversations for context, and sends a proactive message from the workspace. Conduit links contacts to their tickets, calls, reservations, and appointments, so the outreach is grounded in the full history rather than a cold message.

Example prompt: Read the contact with GET `/v1/contacts/{id}`, then send a follow-up with POST `/v1/contacts/{id}/messages`

### Support Performance Monitoring

An operations agent polls Conduit for the workspace automation and resolution rates to track how much support load the AI agents are carrying. The insights endpoints return the metrics as structured JSON the agent can chart or alert on, without pulling the full conversation log.

Example prompt: Call GET `/v1/insights/automation_rate` and GET `/v1/insights/resolution_rate` to report current support automation

### AI Agent Integration via Jentic

An AI agent reaches Conduit through Jentic without handling the bearer token itself. It searches by intent, receives the matching operation and its input schema, and executes the call with the credential injected at run time. This lets a builder wire an external agent into Conduit's ticketing and knowledge-base operations without managing the token in code.

Example prompt: Search Jentic for 'resolve a support ticket', load the POST `/v1/tickets/{id}/resolve` schema, and execute it with Jentic-managed credentials

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| POST | `/v1/agents/{id}/run` | Run a V2 chat agent |
| GET | `/v1/tickets` | List support tickets |
| POST | `/v1/tickets/{id}/messages` | Send a message to a ticket |
| POST | `/v1/tickets/{id}/resolve` | Resolve a ticket |
| POST | `/v1/kb/search` | Search the knowledge base |
| POST | `/v1/contacts/{id}/messages` | Send a proactive message to a contact |
| GET | `/v1/insights/automation_rate` | Get the workspace automation rate |

## Key resources

- **Agents** — Create, run, simulate, and configure AI chat agents via `/v1/agents`
- **Tickets** — Read, message, resolve, close, and reopen support tickets via `/v1/tickets`
- **Conversations** — List, search, and read messages for customer conversations via `/v1/conversations`
- **Knowledge base** — Search and manage knowledge nodes via `/v1/kb/search` and `/v1/kb/nodes`
- **Contacts** — Manage contact records and send proactive messages via `/v1/contacts`

## 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:** 60 / 100
- **Maturity:** AI-Aware
- **Dimensions:**
  - Foundational Compliance: 73 / 100
  - Developer Experience & Jentic Compatibility: 64 / 100
  - AI-Readiness & Agent Experience: 45 / 100
  - Agent Usability: 67 / 100
  - Security: 60 / 100
  - AI Discoverability: 78 / 100
- **View full report:** https://jentic.com/apis/conduit.ai/conduit/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 <openapi-url>
```

## Why Jentic

- **Setup:** Wiring Conduit by hand means managing its bearer token, scoping write access correctly, and threading the token through every ticket, conversation, and knowledge-base call. Through Jentic you install once, import Conduit from the API Directory, store the token once, and your agent calls it.
- **Permission scoping:** Conduit puts the ticket id in the URL path (`/v1/tickets/{id}/...`), so a rule can pin your agent to the ticket operations and leave agent creation, workspace settings, and portal templates out of reach. You choose which operations it may call, so destructive ones like knowledge-node deletion are not included unless you add them.
- **Credential handling:** Your Conduit 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 'resolve a support ticket' or 'search the knowledge base', and Jentic returns the matching Conduit operation with its input schema so the agent calls the right endpoint without browsing the reference docs.

## Related APIs

- **Intercom** — Customer messaging and support suite with its own AI resolution bot
- **Zendesk** — Established ticketing and helpdesk platform with broad integrations
- **Help Scout** — Shared-inbox help desk with conversations and docs

## FAQ

### What authentication does the Conduit API use?

The Conduit API authenticates with a bearer token in the Authorization header, per its OpenAPI spec, where read and write endpoints require a token with write access. Through Jentic the token is stored encrypted by your own Jentic One instance and injected at execution time, so it never reaches the agent.

### Can I resolve support tickets with the Conduit API?

Yes. A ticket exposes its message history and resolve, close, and reopen actions, so an agent can post a reply and then resolve the ticket in one flow. Ticket notes can also be created and updated for internal context that customers do not see.

### What are the rate limits for the Conduit API?

The OpenAPI spec does not publish specific rate-limit numbers; check https://docs.conduit.ai for the current limits on your plan. The billing usage endpoint reports your workspace consumption so an agent can gauge how much headroom it has before a batch of runs.

### Is there a Conduit MCP server?

You don't need an MCP server to give your agent Conduit. Jentic connects it directly from the API Directory: import it, store your credential once, and your agent calls the ticketing, conversation, and knowledge-base operations. Nothing extra is loaded into the agent's context until an operation is actually invoked.

### Can I limit what my agent is allowed to do with the Conduit API?

Yes. Conduit puts the ticket id in the URL path, so write a rule that allows only the ticket message and resolve operations under `/v1/tickets`, and the agent can work tickets while leaving agent creation and workspace settings untouched. Every call runs through your own Jentic One instance and is logged for you to review.

### How do I resolve a ticket with the Conduit API through Jentic?

Search Jentic for 'resolve a support ticket', and it returns the Conduit resolve operation with its input schema. Load the schema and execute POST `/v1/tickets/{id}/resolve` with the credential injected at run time. To run it on your own infrastructure, install Jentic One from its GitHub repo.