> ## Documentation Index > Fetch the complete documentation index at: https://docs.gloo.com/llms.txt > Use this file to discover all available pages before exploring further. # Agents > Built-in agents for common workflows, plus how to create custom agents with any Gloo AI model. GlooCode uses a multi-agent architecture. Each agent is a specialized mode optimized for a different type of work — planning, implementing, researching, or general coding. You can switch between agents mid-session, or let GlooCode select the right one automatically. ## Built-in Agents GlooCode ships with four agents that cover the most common development workflows: | Agent | Command | What it does | | --------- | ---------- | ----------------------------------------------------------------------------------------------------- | | `plan` | `/plan` | Thinks through architecture, design decisions, and implementation strategy before any code is written | | `build` | `/build` | Focused implementation — writes code, runs tests, orchestrates multi-step changes, and commits | | `explore` | `/explore` | Researches codebases, reads documentation, explores APIs, and gathers context | | `general` | `/general` | The default agent — handles most coding tasks without needing to switch | ### How to Switch Agents Type the agent command (e.g., `/plan`) in the TUI to switch. You can switch as many times as you want within a session. Your conversation history carries over, so the new agent has full context of what you've been working on. ### When to Use Each Agent **Use `plan` when** you're starting something complex — a new feature, a migration, a refactor that touches many files. It will think through the approach, identify risks, and lay out a step-by-step plan before any code is written. **Use `build` when** you know what you want and need focused execution. You have a plan (or the task is straightforward enough to not need one), and you want code written, tests passing, and changes committed. **Use `explore` when** you're unfamiliar with a codebase or need to understand how something works before changing it. It reads files, traces call chains, and summarizes what it finds. **Use `general` when** the task is self-contained and doesn't need a specialized approach. This is the default — if you're not sure which agent to use, start here. ## Custom Agents You can define your own agents that use any [model available through Gloo AI](/api-guides/supported-models). Custom agents let you create specialized tools for your team's specific workflows — code reviewers, QA testers, documentation writers, security auditors, or anything else. Custom agents are configured in `~/.config/gloocode/gloocode.json`: ```json theme={null} { "$schema": "https://gloocode.ai.gloo.com/config.json", "agent": { "": { "mode": "subagent", "model": "", "description": "What this agent does.", "prompt": "System prompt that defines the agent's behavior.", "permission": { "glob": "allow", "grep": "allow", "read": "allow", "bash": "allow", "edit": "allow" } } } } ``` ### Configuration Fields | Field | Required | Description | | ------------- | -------- | ----------------------------------------------------------------------------------------------------------------------------------- | | `mode` | Yes | Always `"subagent"` for custom agents | | `model` | Yes | A Gloo AI model ID (e.g., `gloo-anthropic-claude-sonnet-4.5`). See [Supported Models](/api-guides/supported-models) for all options | | `description` | Yes | A short summary shown when listing agents — helps you remember what each one does | | `prompt` | Yes | The system prompt that defines how the agent behaves, what it focuses on, and how it should respond | | `permission` | No | Fine-grained control over what the agent can do. Defaults to all allowed if omitted | ### Example: Code Review Agent This agent can read code and run commands but **cannot edit files** — useful for getting an independent review of your changes before committing. ```json theme={null} { "$schema": "https://gloocode.ai.gloo.com/config.json", "agent": { "review": { "mode": "subagent", "model": "gloo-anthropic-claude-sonnet-4.5", "description": "Post-implementation code reviewer.", "prompt": "Review the diff for bugs, security issues, and performance problems. Be specific about what's wrong and why. Your verdict must be APPROVED, CHANGES REQUESTED, or BLOCKED.", "permission": { "glob": "allow", "grep": "allow", "read": "allow", "bash": "allow", "edit": "deny" } } } } ``` Once saved, switch to this agent with `/review` in the TUI. ### Example: QA Agent This agent tries to break your implementation by running tests and checking edge cases — also read-only so it can't fix what it finds. ```json theme={null} { "$schema": "https://gloocode.ai.gloo.com/config.json", "agent": { "qa": { "mode": "subagent", "model": "gloo-openai-gpt-5.5-pro", "description": "Pre-commit verification agent.", "prompt": "Try to break the implementation. Run tests, check edge cases, verify API contracts. Report what you find with severity levels. Your verdict must be PASS, FAIL, or PARTIAL.", "permission": { "glob": "allow", "grep": "allow", "read": "allow", "bash": "allow", "edit": "deny" } } } } ``` ### Permission Reference Permissions control what tools a custom agent can use. Set each to `"allow"` or `"deny"`. If you omit the `permission` block entirely, all permissions default to `"allow"`. | Permission | What it controls | | ----------- | ------------------------------------------------------------ | | `glob` | Search for files by name or pattern | | `grep` | Search inside file contents | | `read` | Read file contents | | `bash` | Execute shell commands (tests, builds, scripts) | | `edit` | Create or modify files | | `lsp` | Language server features (go-to-definition, find references) | | `webfetch` | Fetch content from URLs | | `websearch` | Search the web | | `skill` | Use skills and plugins | | `task` | Create and manage task lists | | `question` | Ask you clarifying questions | **Best practice:** Read-only agents (reviewers, auditors) should set `"edit": "deny"` to prevent accidental modifications. This is especially useful when you want an independent second opinion — the agent can analyze but not change anything. ### Available Models Custom agents can use any model available through Gloo AI. Model IDs follow the format `gloo--`. Here are some common choices: | Use case | Suggested model | Why | | ---------------------------- | ---------------------------------- | ------------------------------------------------ | | Code review, planning | `gloo-anthropic-claude-sonnet-4.5` | Strong reasoning, good at catching subtle issues | | Fast iteration, simple tasks | `gloo-anthropic-claude-haiku-4.5` | Fast and cost-effective for straightforward work | | Alternative perspective | `gloo-openai-gpt-5.5-pro` | Different model family for diverse review | | Cost-sensitive bulk work | `gloo-google-gemini-2.5-flash` | Low cost per token for high-volume tasks | See [Supported Models](/api-guides/supported-models) for the full list. ## Related Already have custom rules or agents in another tool? Migrate them to GlooCode. Install GlooCode and connect it to your Gloo AI account. Use the same Gloo AI credentials with the OpenAI Python and Node.js SDKs.