Skip to main content
An Agent combines instructions, a model, and tools for one kind of work. Make one when you keep giving the same instructions, or when a job needs a particular set of tools. Manage your Agents in Settings → Agent List.

​
Create an Agent

  1. In Settings → Agent List, click New Agent, or run Create Agent from SmartBar.
  2. Give it a clear name and purpose.
  3. Under Instructions, describe what it should do, what input it needs, and how it should present results.
  4. Choose its model, and set up the tools and context sources it needs.
  5. Save, start a new chat with it, and test a small, typical task.
Example instruction:

​
Choose tools

Give the Agent only the tools its job needs: The model must support the tools and input types you choose. Tools also need their own credentials, access, and setup, and an instruction can’t grant a permission the Agent doesn’t have.

​
Approve actions

The Permissions chip next to the run mode sets when an Agent stops to ask before it acts. It appears in Agent mode in the Chat Window, SmartBar, the App Sidebar, and the Dynamic Island. All of them change the same setting for that Agent, so a change in one shows in the others. Reading files, searching, and looking at the screen never ask.

​
Where the question appears

The question appears in the window the chat runs in. If that window isn’t showing the chat (for example, it’s hidden, or it’s the Inline Panel), the question moves to the Dynamic Island. If nobody answers there within 30 seconds, your paired iPhone gets a reminder notification, even when the chat’s window is still open behind the app you’re using. If you close the island’s card without answering, click the waiting step in the island’s chat to see the options again. Questions the Agent asks you work the same way.

​
Answer it

  • Yes, allow once lets this one action run.
  • Yes, and don’t ask again allows it for the rest of that chat, and is remembered even after Enconvo restarts.
  • Or type what the Agent should do instead.
While a chat waits, it shows Waiting for approval instead of running in the Chat Window’s list, SmartBar’s recent chats, and the Dynamic Island. The iPhone app’s chat list marks it Needs approval. Subagents ask in their parent chat, and the question says that a subagent is asking and which task it’s working on. Scheduled tasks, workflows, and IM channels run without asking, because nobody is at the window to answer.

​
Subagents

In Agent mode, an Agent can split a task into parts and start up to six subagents that work at the same time, each in its own conversation. It answers once they report back. It does this only when splitting clearly helps, such as several independent searches, or when you ask. If an AI error such as a rate limit stops a subagent partway, the Agent still gets what it found so far, and can let it continue once the error clears.

​
Follow and talk to subagents

A card in the chat lists each subagent by its task, with an icon that shows whether it’s running, waiting for your approval, done, or failed. A running subagent’s row also shows what it’s doing right now.
  • Click a row to open that subagent’s conversation. Its Workspace panel and @ file search show the working folder of the chat that started it. In SmartBar and the Dynamic Island, its work opens in place.
  • Type in a subagent’s conversation to talk to it. It answers you there, can ask you questions, asks for approval in that conversation, and keeps working in the same folder. The Agent that started it waits while it answers you, and then sees its newest answer.
  • In the Chat Window’s sidebar, a chat’s subagents are listed under it, and a button on the chat’s row folds the list.

​
Subagent roles

Every subagent starts in a role: To choose one yourself, type @ in an Agent-mode chat and pick it from the Subagents group, which lists the project’s helpers, Explorer, and a general Subagent. The Agent starts that subagent for your message, writes it the task, and answers once it reports back. Your own Agents stay in the menu’s Agents group. Describe each Agent clearly in its settings: the Agent that splits the work reads those descriptions to decide which one fits a task.

​
Project helpers

Project helpers use the same files as Claude Code subagents and Codex agent roles, so a repository that already has them needs no extra setup.
  • A Claude Code file (Markdown with name and description in its frontmatter) is read-only when it sets permissionMode: plan, or lists only reading tools such as Read, Grep, and Glob.
  • A Codex file (TOML with description and developer_instructions) is read-only with sandbox_mode = "read-only".
  • When the provider doesn’t offer the helper’s model, the helper uses the chat’s model instead.
A Claude Code file’s tools and disallowedTools decide which of the Agent’s tools the helper may use. Enconvo matches each name to its own tool: With no tools line, the helper has every tool except the ones disallowedTools names. A call outside the list isn’t run, and the helper is told which tools it has. Every helper can still load skills and ask you questions. In an .enconvo/agents file, you can also list Enconvo tools by their API path, such as browser-use/*.

​
Ways to reach an Agent

​
Defaults and continuity

Check the model defaults before you change an Agent that several chats or channels use. Set up Memory for facts the Agent should reuse, and keep documents in a knowledge base instead of putting everything into the instructions.

​
Troubleshooting

Chat Window

Hold conversations with your Agents.

Skills

Teach an Agent a reusable procedure.

MCP Servers

Connect external services as tools.

Hooks

Run your own commands when an Agent acts.