Visual agent and workflow platform

How to use
Coze.

Composing and publishing AI agents and workflows with prompts, models, knowledge, variables, plugins, APIs, and channel integrations without building every orchestration surface from scratch.

Open Coze ↗ Official download ↗
What it isvisual platform and API for building, evaluating, publishing, and operating AI agents, workflows, knowledge, plugins, tools, and multi-channel applications Workflows2 UpdatedSeptember 2026

Coze is a visual platform and API for building, evaluating, publishing, and operating AI agents, workflows, knowledge, plugins, tools, and multi-channel applications. Composing and publishing AI agents and workflows with prompts, models, knowledge, variables, plugins, APIs, and channel integrations without building every orchestration surface from scratch. This guide covers the whole path in one place: official access, a first session that produces something reviewable, the checks that make output trustworthy, and the permissions worth limiting before you connect real work.

Troiana principle

AI should make the work easier to inspect. If the workflow removes the source, the owner, or the review step, redesign the workflow.

01

Access & setup

Find, install, and sign in to Coze

Get into the official Coze experience with the right account and a setup you understand.

  1. 01

    Start at https://www.coze.com/ and confirm the domain before entering account or payment information.

  2. 02

    Availability: Coze is primarily a browser platform with workspaces, visual agent and workflow builders, API management, personal access tokens and OAuth applications, publishing channels, and developer APIs.

  3. 03

    A supported Coze account and workspace, approved models and plugins, scoped server-side authentication, a permissioned knowledge corpus, end-user notices and consent, workflow tests, usage credits, monitoring, and release ownership.

  4. 04

    Sign in with the account you intend to keep using, then review plan, data, notification, and permission settings.

  5. 05

    Run one low-risk test task before connecting sensitive files, repositories, or workspace data.

  • Use official download pages.
  • Review permissions during setup.
  • Keep installers and applications updated.
02

First session

Your first useful Coze session

Learn the interaction loop using a small task with a clear outcome.

  1. 01

    Define the agent's users, purpose, allowed data, knowledge, models, tools, channels, side effects, consent, quality and safety gates, token budget, retention, and human escalation.

  2. 02

    Create an isolated workspace project, add the minimum approved knowledge and plugins, use explicit workflow inputs and outputs, and keep personal access tokens and OAuth secrets out of prompts and browser code.

  3. 03

    Build a bounded draft with tool allowlists, schemas, timeouts, branches, stopping conditions, citations or abstention where needed, and human approval before consequential external actions.

  4. 04

    Evaluate normal, ambiguous, no-answer, unsafe, injection, permission, tool-failure, duplicate-action, cost, and channel cases, then publish to a limited audience with logs, feedback, revocation, and rollback.

  • State the outcome before the background.
  • Provide the real source material.
  • Review the result before expanding the task.
03

Quality control

Check the quality of Coze output

Establish that an autonomous run did the right thing, not merely that it finished.

  1. 01

    Define what the run should achieve and what it must never touch before granting it a single tool.

  2. 02

    Read the full execution trace: which tools were called, with what arguments, and in what order.

  3. 03

    Verify the side effects directly in the target system rather than trusting the agent's own report of success.

  4. 04

    Confirm failures surfaced as failures — a silent retry loop or a swallowed error is more dangerous than a crash.

  5. 05

    Re-run the same task and compare: an agent that behaves differently across identical runs is not yet production-ready.

  • Verify side effects in the system of record, not in the agent's summary.
  • Require human approval for any irreversible or outward-facing action.
  • Log every tool call so a run can be reconstructed afterwards.
04

Privacy & permissions

Use Coze safely

Bound what an autonomous system can reach before you let it run unattended.

  1. 01

    Enumerate every tool, credential, and system the agent can reach, and remove the ones it does not need.

  2. 02

    Require explicit human approval for irreversible actions: sending, publishing, paying, deleting, or deploying.

  3. 03

    Run against non-production data until behaviour is predictable across repeated runs.

  4. 04

    Set hard limits on spend, iterations, and runtime so a failure loop cannot run unbounded.

  5. 05

    Treat anything the agent reads from the web or a document as data, never as instructions it may follow.

  • Coze agents can combine third-party models, plugins, APIs, knowledge, and publishing channels, so every component adds separate data terms and trust boundaries. Coze warns against exposing personal access tokens in browser code, places end-user notice and consent duties on developers, and requires lawful handling of plugin requests and protected information. Minimize shared data, review every third party, scope credentials, validate and reauthorize tool calls, make writes idempotent, require approval for consequential actions, publish privately first, and support correction, export, revocation, and deletion.
  • Follow your organisation's approved-use policy.
  • Never treat fluent output as authorization to act.
05

Core workflows

Step-by-step ways to use Coze for the work it does best.

Each workflow below has its own outcome, steps and review checkpoints.

01 WorkflowBuild and evaluate a governed Coze agent

Compose prompts, knowledge, workflows, models and plugins behind explicit scope, evidence, permission, cost, safety, and release gates.

 02 WorkflowPublish a secure Coze API and plugin integration

Separate user identity, Coze credentials, plugin service accounts, consent, request validation, execution, and audit across the full integration chain.

Workflow 01

Build and evaluate a governed Coze agent

Compose prompts, knowledge, workflows, models and plugins behind explicit scope, evidence, permission, cost, safety, and release gates.

Outcome: Publish an agent that answers or acts within a documented boundary and fails safely when knowledge, authorization, tools, or models are uncertain.

  1. 01

    Define users, purpose, prohibited uses, data classes, knowledge owners, models, tools, channels, output and citation rules, side effects, budget, retention, escalation, and rollback.

  2. 02

    Create an isolated project, ingest minimal approved knowledge with source and expiry metadata, choose reviewed models and plugins, and keep prompts and variables free of secrets.

  3. 03

    Build explicit workflow inputs, branches, schemas, timeouts, retries and stopping conditions; restrict tool destinations and permissions, reauthorize resources, and require approval before consequential actions.

  4. 04

    Create evaluations for normal, ambiguous, stale, conflicting, no-answer, multilingual, unsafe, injection, knowledge-poisoning, permission, plugin-failure, duplicate-action and token-exhaustion cases.

  5. 05

    Gate publication on results, publish privately to a limited audience, monitor traces, feedback, cost and tool calls, refresh reviewed knowledge, revoke compromised integrations, and retain a prior version for rollback.

What good use looks like

  • Use evaluations before publication.
  • Require evidence or abstention.
  • Approve consequential tool calls.

Workflow 02

Publish a secure Coze API and plugin integration

Separate user identity, Coze credentials, plugin service accounts, consent, request validation, execution, and audit across the full integration chain.

Outcome: Expose an agent or tool without leaking personal tokens, confusing tenant authority, mishandling end-user data, or allowing replayed side effects.

  1. 01

    Map end users, developer workspace, agent, OAuth app or personal token, plugin API, third parties, data fields, permissions, notices, consent, side effects, quotas, retention, and incident owners.

  2. 02

    Prefer OAuth or dedicated scoped service credentials for production, keep every secret server-side, separate environments and tenants, set expiry and rotation, and never share one personal token across users.

  3. 03

    Validate signatures and schemas, canonicalize identifiers and URLs, bind requests to the authenticated user and tenant, enforce authorization at the plugin API, and reject unknown fields and destinations.

  4. 04

    Make writes idempotent, require confirmation for messages, publication, purchases and account changes, use timeouts and rate limits, redact logs, and return bounded responses with no credentials or excessive records.

  5. 05

    Test token leak and revocation, forged identity, tenant crossover, prompt injection, SSRF, replay, duplicate calls, timeouts, credit exhaustion, consent withdrawal, deletion and rollback before canarying.

What good use looks like

  • Keep all Coze tokens server-side.
  • Authorize at the plugin service.
  • Make external writes idempotent.
06

Official references

Check the current product documentation.

Features, plan limits, availability, and data controls change. These official pages are the starting points used for this guide.

Explore the Troiana AI Hub →