> ## Documentation Index
> Fetch the complete documentation index at: https://docs.argentos.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Tools Overview

> The 50+ tools available to ArgentOS agents — file operations, web research, memory, tasks, knowledge, communication, and more.

## What Are Tools?

Tools are how an ArgentOS agent takes action. When the agent needs to run a command, search the web, recall a memory, or manage a task, it invokes a tool. Each tool is defined as a JSON schema, validated at runtime, and governed by the agent's intent policies and sandbox mode.

The agent selects tools automatically based on conversational context. You do not need to tell it which tool to use — ask for what you need and the agent routes to the right capability.

## Tool Catalog

<CardGroup cols={3}>
  <Card title="File & Code" icon="file-code">
    `read`, `write`, `edit`, `bash`, `glob`, `grep` — full filesystem and shell access
  </Card>

  <Card title="Web & Research" icon="globe">
    `browser`, `web_search`, `web_fetch`, `firecrawl` — browse, search, scrape
  </Card>

  <Card title="Memory" icon="brain">
    `memory_store`, `memory_recall`, `memory_timeline` — persistent agent memory
  </Card>

  <Card title="Tasks" icon="list-check">
    `tasks` — 15+ actions for full task lifecycle management
  </Card>

  <Card title="Knowledge & Library" icon="book">
    `knowledge_search`, `knowledge_ingest`, `knowledge_collections` — RAG library with ACL
  </Card>

  <Card title="Communication" icon="paper-plane">
    `outbound_message`, `channel_send` — multi-channel delivery
  </Card>

  <Card title="Media & Display" icon="image">
    `sag`, `image_gen`, `doc_panel`, `canvas` — speech, images, document panels
  </Card>

  <Card title="System & Config" icon="gear">
    `os_docs`, `health`, `config`, `gateway`, `backup` — system operations
  </Card>

  <Card title="Marketplace" icon="store">
    `marketplace_search`, `marketplace_install`, `marketplace_publish` — package management
  </Card>
</CardGroup>

### File & Code

| Tool    | Description                              |
| ------- | ---------------------------------------- |
| `read`  | Read file contents from disk             |
| `write` | Create or overwrite a file               |
| `edit`  | Apply targeted edits to an existing file |
| `bash`  | Execute shell commands and scripts       |
| `glob`  | Find files by pattern matching           |
| `grep`  | Search file contents with regex          |

### Web & Research

| Tool         | Description                                                  |
| ------------ | ------------------------------------------------------------ |
| `browser`    | Automate web browsing — navigate, click, extract content     |
| `web_search` | Search the web via configured search provider                |
| `web_fetch`  | Fetch a URL and return its content (with Firecrawl fallback) |
| `firecrawl`  | Deep web scraping and crawling via Firecrawl API             |

### Memory

| Tool              | Description                                                      |
| ----------------- | ---------------------------------------------------------------- |
| `memory_store`    | Save observations, facts, or context to persistent memory (MemU) |
| `memory_recall`   | Search and retrieve from memory using hybrid FTS + vector search |
| `memory_timeline` | Browse memory chronologically with filtering                     |

### Tasks

| Tool    | Description                                                                                                                                         |
| ------- | --------------------------------------------------------------------------------------------------------------------------------------------------- |
| `tasks` | Full task management — create, list, start, complete, block, prioritize, assign dependencies, organize into projects. 15+ actions in a single tool. |

### Knowledge & Library

| Tool                    | Description                                                         |
| ----------------------- | ------------------------------------------------------------------- |
| `knowledge_search`      | Search the RAG knowledge library (hybrid vector + BM25, ACL-scoped) |
| `knowledge_ingest`      | Ingest documents into knowledge collections (PDF, DOCX, XLSX, text) |
| `knowledge_collections` | Manage collections, grants, and access control                      |

### Communication & Channels

| Tool               | Description                                                          |
| ------------------ | -------------------------------------------------------------------- |
| `outbound_message` | Send messages through the outbound pipeline (multi-channel delivery) |
| `channel_send`     | Send to a specific channel (Telegram, Discord, Slack, WhatsApp)      |

### Media & Display

| Tool        | Description                                                      |
| ----------- | ---------------------------------------------------------------- |
| `sag`       | Speech audio generation (text-to-speech via ElevenLabs or local) |
| `image_gen` | Generate images via configured provider                          |
| `doc_panel` | Open documents in the dashboard slide-out panel                  |
| `canvas`    | Send content to external node device screens                     |

### System & Configuration

| Tool      | Description                                                   |
| --------- | ------------------------------------------------------------- |
| `os_docs` | Read ArgentOS documentation and guides                        |
| `health`  | Check system health, gateway status, and service connectivity |
| `config`  | Read or update agent configuration                            |
| `gateway` | Interact with the gateway control plane                       |
| `backup`  | Trigger or manage Phoenix backups (local/S3/R2)               |

### Marketplace

| Tool                  | Description                             |
| --------------------- | --------------------------------------- |
| `marketplace_search`  | Search the AOS Marketplace for packages |
| `marketplace_install` | Install a package from the marketplace  |
| `marketplace_publish` | Publish a package to the marketplace    |

### Automation & Scheduling

| Tool       | Description                                           |
| ---------- | ----------------------------------------------------- |
| `cron`     | Manage scheduled jobs (create, list, enable, disable) |
| `webhooks` | Register and manage inbound webhook triggers          |

### Project Management

| Tool        | Description                                                                                               |
| ----------- | --------------------------------------------------------------------------------------------------------- |
| `specforge` | Kick off the SpecForge project intake workflow — type gating, interviews, drafts, approval, and execution |

## How Tools Activate

Tools are selected by the LLM based on context. The flow:

<Steps>
  <Step title="LLM returns a tool call">
    The LLM returns a `tool_use` block with the tool name and input parameters.
  </Step>

  <Step title="Schema validation">
    The agent runtime validates input against the tool's JSON schema.
  </Step>

  <Step title="Policy and sandbox check">
    Intent policies and sandbox mode are checked.
  </Step>

  <Step title="Tool executes">
    The tool handler executes and returns a result.
  </Step>

  <Step title="Result feeds back">
    The result feeds back to the LLM as a `tool_result` block.
  </Step>
</Steps>

You can guide tool selection by being specific ("search my memory for..." or "create a task to..."), but explicit tool names are never required.

## Tool Governance

The [Intent System](/intents) controls which tools an agent can use. Intent policies define `allowedActions` and `requiresHumanApproval` lists that restrict tool access at three levels:

* **Global** — baseline restrictions for all agents
* **Department** — narrowed per department (e.g., Support agents cannot use `bash`)
* **Agent** — per-agent overrides (most restrictive wins)

<Warning>
  Policies are monotonic: child levels can only tighten constraints, never loosen them. If Global allows a tool but the Department policy removes it, no agent in that department can use it.
</Warning>

## Sandbox Modes

Tool execution is gated by the agent's security mode:

| Mode             | Behavior                                                                       |
| ---------------- | ------------------------------------------------------------------------------ |
| **Unrestricted** | All tools available, no approval needed                                        |
| **Sandboxed**    | Destructive tools (file writes, bash, outbound messages) require user approval |
| **Locked**       | Only safe, read-only tools (memory recall, task listing, search) are available |

See [Sandboxing](/security/sandboxing) for configuration details.

## Extending Tools

**Plugins** can register new tools at startup. See [Plugins](/tools/plugins).

**Skills** bundle tools with prompts into loadable capability packages. See [Skills](/tools/skills).