Agent Tools Overview

Master reference for every tool your Comis agents can use

Comis ships with 73 built-in agent tools spanning file operations, web access, messaging, scheduling, media processing, browser automation, context management, interactive terminal sessions, and administration. This page is the master reference — every tool listed with its category, description, and link to detailed documentation.

When an agent receives a message, it decides which tools to use based on what you asked it to do. Need a daily briefing? The agent uses the cron tool. Want to send a message to another channel? It uses message. Need to analyze an image someone sent? It reaches for image_analyze.

The tools listed below are everything your agents have at their disposal. Each tool is linked to its detailed guide page where you will find parameters, examples, and tips for effective use.

Understanding Tool Categories

Tools fall into three tiers:

Tier	Count	Description
Built-in tools	73	Core capabilities shipped with Comis — file operations, shell commands, web search, browser automation, messaging, scheduling, sessions, media, infrastructure, context, interactive terminal sessions, supervisor admin, and platform actions. Available on every agent unless restricted by tool policy.
MCP tools	Configurable	External Model Context Protocol servers — Anthropic’s reference servers, community-built MCP servers, and your own. Discovered at runtime and qualified with mcp:{server}/{tool} names. See MCP Management.
Custom skills	Unlimited	Markdown-defined “mini-programs” with optional tool restrictions and bin/env requirements. Comis ships 7 built-in skills; you can author your own. See Skills.

Tool access is controlled by tool policy. By default, agents have access to all tools, but you can restrict access per agent using profiles (like coding or messaging) or fine-grained allow/deny lists.

All tool names use snake_case — for example, web_search, sessions_spawn, and memory_store. When referencing tools in configuration files (such as tool policy allow/deny lists), use these exact names.

Master Reference Table

Every tool your agents can use, organized by category. Click the guide link in any row to jump to the full documentation for that tool.

Files & Code (8 tools)

File and code tools are always available unless explicitly denied by tool policy. They can be individually toggled in config.yaml under the builtinTools section.

Tool	Description	Guide
read	Read file contents with line numbers and pagination	Built-in Tools
write	Create or overwrite files, auto-creates parent directories	Built-in Tools
edit	Structured insert/replace/delete edits on existing files	Built-in Tools
notebook_edit	Cell-level Jupyter notebook editing	Built-in Tools
grep	Search file contents using ripgrep (auto-provisioned)	Built-in Tools
find	Find files using fd (auto-provisioned) or fallback	Built-in Tools
ls	List directory contents	Built-in Tools
apply_patch	Apply unified diffs to files	Built-in Tools

Process & Execution (2 tools)

Tool	Description	Guide
exec	Sandboxed shell execution with restricted-binary policy	Built-in Tools
process	Background process lifecycle (spawn, kill, status, logs)	Built-in Tools

Terminal (Interactive Shell)

Interactive terminal sessions for driving allowlisted binaries. Requires an operator-configured allowEntries allowlist — without a matching entry the tool rejects. Requires bubblewrap sandbox on Linux. See exec sandbox. Full reference: the Terminal Driver guide; for driving Claude Code & Codex as first-class coding agents (config + how it works), see Coding CLIs; to hand the agent a design doc and have it build the whole thing via the GSD spec-driven workflow, see Autonomous Builds.

Tool	Description	Guide
terminal_session_create	Start an interactive terminal session driving an allowlisted binary.	Built-in Tools
terminal_session_read	Read buffered output from an active session (accumulated since last read).	Built-in Tools
terminal_session_list	List all active terminal sessions for the current agent.	Built-in Tools
terminal_session_kill	Terminate an active terminal session and free its resources.	Built-in Tools
terminal_session_send_text	Write text to a terminal session’s stdin (optionally submit with Enter after a settle).	Built-in Tools
terminal_session_send_key	Send named key chords (e.g., Enter, Ctrl-C, S-Tab) to a terminal session.	Built-in Tools
terminal_session_resize	Resize a terminal session’s pseudo-TTY (cols x rows).	Built-in Tools
terminal_session_wait	Wait for a terminal session to produce matching output or a timeout.	Built-in Tools
terminal_session_status	Registered with final schema; execute rejects with [not_implemented] (stub).	Built-in Tools

Web & Browser (3 tools)

Tool	Description	Guide
web_search	Web search across 8 providers (Brave, DuckDuckGo, Exa, Grok, Jina, Perplexity, SearXNG, Tavily) with optional fallback chain	Web Tools
web_fetch	Fetch a URL and extract readable content via Readability (SSRF-guarded)	Web Tools
browser	Headless Chromium automation — 16 actions, 11 act sub-kinds	Browser

Scheduling

Scheduling tools let agents create automated tasks that run on a timer. Agents can set up recurring cron jobs, one-time reminders, and manage their own heartbeat — a periodic wake-up that triggers routine maintenance or check-in tasks.

Tool	Description	Guide
cron	Create and manage scheduled jobs	Scheduling
heartbeat_manage	Configure agent heartbeat timing and triggers	Scheduling

Messaging

The message tool works across all connected channels — Discord, Telegram, Slack, WhatsApp, and every other supported platform. Agents can send new messages, reply to specific messages, add emoji reactions, edit or delete messages, fetch message history, and send file attachments.

Agents can also send messages to channels they are not currently conversing in, making the message tool ideal for notifications, alerts, and cross-channel coordination.

Tool	Description	Guide
message	Send, reply, react, edit, delete, fetch, and attach messages	Messaging
notify_user	Send proactive notifications, alerts, and reminders to the user	Messaging

Sessions & Sub-Agents

Session tools let agents manage their conversations, communicate with other agents, and spawn sub-agents for specialized tasks. For example, a lead agent can spawn a researcher sub-agent and a writer sub-agent to work in parallel.

Tool	Description	Guide
session_search	Search session history by query	Sessions
session_status	Get the current session status	Sessions
sessions_history	Retrieve session history entries	Sessions
sessions_list	List all sessions for the current agent	Sessions
sessions_send	Send a message into another session in 3 modes (fire-and-forget, wait, ping-pong)	Sessions
sessions_spawn	Spawn a sub-agent session with workspace isolation, artifact refs, and an optional objective	Sessions
subagents	Manage running sub-agents (list, kill, steer)	Sessions

Pipelines (Execution Graphs)

The pipeline tool defines and runs multi-agent execution graphs (DAGs). Nodes run in parallel when independent, or wait for upstream nodes via depends_on. Comis ships 7 built-in node type drivers — agent, debate, vote, refine, collaborate, approval-gate, and map-reduce.

Tool	Description	Guide
pipeline	Define, execute, save, load, list, delete, monitor, cancel, and read outputs from execution graphs (9 actions)	Pipelines

Media

Media tools process images, audio, video, and documents. Agents can analyze images with vision AI, convert text to speech, transcribe voice messages, describe video content, and extract text from documents like PDFs. These tools require appropriate AI providers to be configured (for example, vision capabilities for image analysis, and STT providers for transcription).

Tool	Description	Guide
image_analyze	Vision AI image analysis (file, URL, base64, attachment)	Media
tts_synthesize	Text-to-speech audio generation	Media
transcribe_audio	Audio and voice transcription to text	Media
describe_video	Video description using vision AI	Media
extract_document	Document text extraction (PDF, CSV, TXT, and more)	Media
image_generate	Generate images from text prompts, delivered to the current channel	Media

Memory

Memory tools give agents access to their long-term memory. Agents automatically build up memory over time, but these tools let them explicitly search, read, and store information. Memory uses hybrid text and vector search for semantic matching, so agents can find relevant memories even with different wording.

Tool	Description	Guide
memory_get	Retrieve a memory entry by key	Sessions
memory_manage	Create, update, or delete memory entries	Sessions
memory_search	Semantic search over agent memory (hybrid text + vector)	Sessions
memory_store	Store a new memory entry (secret-detection guard applied)	Sessions
memory_ask	Grounded natural-language answer over the agent’s recall pipeline. Default-on; disable per-agent with dialectic.enabled: false, or via the memory.enabled: false kill switch.	Sessions

Context (3 tools)

In-session recovery tools over the DAG (LCD) context engine’s lossless store. Available when the agent runs in DAG mode (contextEngine.version: "dag"). They recover detail the summary hierarchy compressed away within this conversation — they are distinct from cross-session recall (memory_search / session_search), which they share no code with, and they never reach other sessions or tenants. All three are never-export (never reachable via the MCP server) and read-only. See Context Management.

Tool	Description	Documentation
ctx_search	Full-text search over THIS conversation’s compressed history (FTS with a LIKE fallback).	Sessions
ctx_inspect	Inspect a summary’s metadata (depth, coverage, time-range) and its immediate children / covered-message count.	Sessions
ctx_expand	Recover (rehydrate) a compressed region back to its underlying messages; large output is written to a session file with a handle.	Sessions

Infrastructure

These platform-management tools let agents inspect or modify the running Comis daemon. The gateway tool drives configuration and the secret store; background_tasks inspects the long-running task queue; mcp_manage reaches into the MCP client manager. See also: browser (listed under the Web and Browser section).

Tool	Description	Guide
gateway	Configuration management and gateway control (11 actions: read, patch, apply, restart, schema, status, history, diff, rollback, env_set, env_list)	Infrastructure
background_tasks	Inspect and manage background task queue (list, get, cancel, read_output)	Infrastructure
mcp_manage	MCP server lifecycle (list, status, connect, disconnect, reconnect)	Infrastructure
mcp_login	MCP provider OAuth login (e.g., for Higgsfield and other MCP services that require authentication)	Infrastructure
get_prompt	Fetch a named prompt from a connected MCP server. Conditional: active only when a connected server advertises prompts.	Infrastructure
list_prompts	List available prompts from connected MCP servers. Conditional: active only when a connected server advertises prompts.	Infrastructure
list_resources	List available resources from connected MCP servers. Conditional: active only when a connected server advertises resources.	Infrastructure
read_resource	Read a specific resource from a connected MCP server. Conditional: active only when a connected server advertises resources.	Infrastructure

Supervisor

Supervisor tools require admin-level trust. They manage the entire platform, not just the current agent’s session. Only agents with the supervisor profile or explicit allow entries can use these tools.

Supervisor tools can modify agents, delete sessions, flush memory, and manage gateway tokens. Grant these tools only to trusted agents that need fleet-wide control. See tool policy for how to restrict access.

Tool	Description	Guide
agents_manage	Agent fleet management (create, get, update, delete, suspend, resume)	Infrastructure
obs_query	Observability and metrics queries (diagnostics, billing, delivery, channels)	Infrastructure
sessions_manage	Session lifecycle management (delete, reset, export, compact)	Infrastructure
channels_manage	Channel adapter management (list, get, enable, disable, restart, configure)	Infrastructure
tokens_manage	Gateway token management (list, create, revoke, rotate)	Infrastructure
models_manage	Model catalog queries (list, test)	Infrastructure
providers_manage	Manage model provider configurations (add, remove, update providers)	Infrastructure
skills_manage	Skill lifecycle management (list, import, delete) with per-agent scope	Infrastructure

See also: heartbeat_manage (listed under the Scheduling section).

Platform Actions

Platform action tools perform moderation and administration tasks specific to each messaging platform. Each tool wraps the native API of its platform, letting agents kick users, manage roles, pin messages, create channels, and more — all from within a conversation.

The available actions vary by platform. See the Platform Actions guide for the full list of actions per platform.

Tool	Actions	Guide
discord_action	19 actions: server moderation (kick, ban, role_add/remove, set_topic, set_slowmode, guild/channel info, threadCreate/List/Reply, channelCreate/Edit/Delete/Move, pin/unpin, setPresence)	Platform Actions
telegram_action	12 actions: chat info & moderation (pin/unpin, poll, sticker, chat_info, member_count, get_admins, set_title, set_description, ban/unban, promote)	Platform Actions
slack_action	12 actions: workspace management (pin/unpin, set_topic, set_purpose, archive/unarchive, create_channel, invite, kick, channel_info, members_list, bookmark_add)	Platform Actions
whatsapp_action	11 actions: group management (group_info, group_update_subject/description, group_participants_add/remove, group_promote/demote, group_settings, group_invite_code, profile_status, group_leave)	Platform Actions

Channel Messaging Capability Matrix

The message tool works across all 10 connected channels, but not every action is supported everywhere. The matrix below summarizes per-channel capabilities — consult the Channels Overview for full per-channel detail.

Channel	Send	Reply	React	Edit	Delete	Fetch History	Threads	Streaming
Discord	Yes	Yes	Yes	Yes	Yes	Yes	Native	Edit
Telegram	Yes	Yes	Yes	Yes	Yes	No	Forum topics	Edit
Slack	Yes	Yes	Yes	Yes	Yes	Yes	Reply chain	Edit
WhatsApp	Yes	Yes (quoted)	Yes	Yes	Own only	No	None	Block
Signal	Yes	Yes (quoted)	Yes	No	Yes	No	None	Block
iMessage	Yes	No	No	No	No	Yes	None	None
LINE	Yes	Yes (push)	No	No	No	No	None	None
IRC	Yes	Mention only	No	No	No	No	None	None
Email	Yes	Yes (headers)	No	No	No	No	Reply chain	None

“Streaming” indicates how live agent output is delivered: Edit means the agent edits a placeholder message as content streams in; Block means the agent buffers and emits complete chunks; None means no streaming support (the agent waits for the full reply).

Quick Access by Use Case

Not sure which tool you need? Start here. The table below maps common tasks to the specific tool and action that handles them. Each row links to the detailed guide page where you can find parameters, examples, and tips.

I want to…	Use this tool	Guide
Send a message to a channel	message (action: send)	Messaging
Send an alert that respects quiet hours	notify_user	Messaging
Schedule a daily task	cron (action: add, schedule_kind: cron)	Scheduling
Set a one-time reminder	cron (action: add, schedule_kind: at)	Scheduling
Run something every N minutes	cron (action: add, schedule_kind: every)	Scheduling
Delegate work to another agent	sessions_spawn	Sessions
Run a multi-step workflow	pipeline (action: execute)	Pipelines
Have multiple agents debate or vote	pipeline (with type_id: debate or vote)	Pipelines
Analyze an uploaded image	image_analyze	Media
Generate an image	image_generate	Media
Transcribe a voice message	transcribe_audio	Media
Speak a reply with TTS	tts_synthesize then message action attach	Media
Search the web	web_search	Web Tools
Read a single web page	web_fetch	Web Tools
Scrape a JavaScript-heavy site	browser (navigate, snapshot, act)	Browser
Kick a user from Discord	discord_action (action: kick)	Platform Actions
Pin the first message of the day	discord_action / telegram_action / slack_action (action: pin)	Platform Actions
Remember something for later	memory_store	Sessions
Read a PDF attachment	extract_document	Media
Check what’s scheduled	cron (action: list)	Scheduling
Toggle a config value	gateway (action: patch then apply)	Infrastructure
Connect a new MCP server	mcp_manage (action: connect)	Infrastructure

Tool Guides

Explore each tool category in detail:

Scheduling

Cron jobs, heartbeat timing, and automated tasks

Messaging

Send, reply, react, edit, and delete messages

Sessions

Sessions, sub-agents, and long-term memory

Pipelines

Multi-agent execution graphs (DAGs) with 7 node types

Media

Image analysis, text-to-speech, transcription, and documents

Infrastructure

Configuration, gateway control, and fleet management

Browser

Headless browser automation and web interaction

Web Tools

Web search across 8 providers and content fetching

Platform Actions

Discord, Telegram, Slack, and WhatsApp moderation