Agents CLI - Pinata Docs

The Agents feature is currently in beta.

The pinata agents commands let you manage your Pinata Agents directly from the terminal. Create agents, start chat sessions, manage skills and secrets, configure channels, and more.

New to Agents?

Learn about the Pinata Agents platform and get started with hosted AI agents.

NAME:
   pinata agents - Interact with AI agents on Pinata

USAGE:
   pinata agents [command [command options]]

COMMANDS:
   list, l          List all agents
   create, c        Create a new agent
   get, g           Get agent details
   delete, d        Delete an agent
   restart, r       Restart an agent
   logs             Get agent logs
   chat, c          Interactive chat with an agent
   exec             Execute a command in an agent container
   skills, sk       Manage agent skills
   secrets, sec     Manage secrets
   channels, ch     Manage agent channels
   devices, dev     Manage agent devices
   snapshots, snap  Manage agent snapshots
   tasks, t         Manage agent cron jobs/tasks
   ports, p         Manage agent port forwarding
   domains, dom     Manage custom domains (beta)
   files            Agent file operations
   templates, tpl   Browse and manage agent templates
   clawhub, hub     Browse and install skills from ClawHub
   config, cfg      Manage agent configuration
   update, up       Manage agent openclaw updates
   versions, ver    List available agent versions
   engines          List the agent engines enabled for this deployment
   auth             Authenticate with a provider and store the credential as a secret
   feedback         Submit feedback or feature request

OPTIONS:
   --help, -h  show help

`list`

NAME:
   pinata agents list - List all agents

USAGE:
   pinata agents list [options]

OPTIONS:
   --help, -h  show help

`create`

NAME:
   pinata agents create - Create a new agent

USAGE:
   pinata agents create [options]

OPTIONS:
   --name string, -n string             Name of the agent (required)
   --description string, -d string      Agent personality description
   --vibe string                        Agent vibe/tagline
   --emoji string                       Agent emoji
   --engine string                      Container engine: openclaw (default) or hermes
   --skill string [ --skill string ]    Skill CIDs to attach (can be specified multiple times)
   --secret string [ --secret string ]  Secret IDs to attach (can be specified multiple times)
   --template string, -t string         Template ID to deploy from (uses template snapshot, skills, and defaults)
   --user-name string                   Owner display name (written into workspace/USER.md)
   --user-email string                  Owner email (written into workspace/USER.md)
   --channels string                    Channel config as JSON (for Hermes agents; set at creation to avoid a restart), e.g. '{"telegram":{"botToken":"...","dmPolicy":"open"}}'
   --help, -h                           show help

--engine selects the container engine. It defaults to openclaw server-side; pass hermes for the opinionated engine. Run pinata agents engines to see which engines your deployment allows. See Concepts → Engine for the differences. Example with template:

# List available templates
pinata agents templates list

# Create agent from template
pinata agents create --name "My IPFS Agent" --template tpchchgg

Example with a Hermes engine and channels: Pass --channels as a JSON object to configure Telegram, Slack, or Discord at creation for engines that support create-time channel bootstrap. This is especially useful for Hermes agents because the channel can come up with the agent and avoid a post-create restart. The per-channel fields (botToken, appToken, dmPolicy, allowFrom) match the Channels API and the channels block in manifest.json.

# Create a Hermes agent
pinata agents create --name "My Hermes Agent" --engine hermes

# Create a Hermes agent with Telegram configured at creation
pinata agents create \
  --name "My Hermes Agent" \
  --engine hermes \
  --user-name "Ada Lovelace" \
  --user-email "ada@example.com" \
  --channels '{"telegram":{"botToken":"123:abc","dmPolicy":"open"}}'

`get`

NAME:
   pinata agents get - Get agent details

USAGE:
   pinata agents get [options] [agent ID]

OPTIONS:
   --help, -h  show help

The response includes the agent’s engine so you can tell OpenClaw, Hermes, and other enabled engines apart.

`delete`

NAME:
   pinata agents delete - Delete an agent

USAGE:
   pinata agents delete [options] [agent ID]

OPTIONS:
   --help, -h  show help

`restart`

NAME:
   pinata agents restart - Restart an agent

USAGE:
   pinata agents restart [options] [agent ID]

OPTIONS:
   --help, -h  show help

`logs`

NAME:
   pinata agents logs - Get agent logs

USAGE:
   pinata agents logs [options] [agent ID]

OPTIONS:
   --help, -h  show help

`chat`

Start an interactive chat session with an agent. Supports multiple output modes for different use cases, and works across supported agent engines.

NAME:
   pinata agents chat - Interactive chat with an agent

USAGE:
   pinata agents chat [options] [agent ID] [optional prompt]

DESCRIPTION:
   Start an interactive chat session with an agent.

   The gateway URL and token are automatically fetched from the agent's configuration.

   Output modes:
     - TTY stdout:     Interactive TUI with markdown rendering
     - Non-TTY stdout: JSONL streaming (machine-readable, default for pipes)
     - --text:         Plain text streaming (simpler alternative to JSONL)
     - --conversation: Multi-turn mode (read messages from stdin line-by-line)

   Examples:
     # Interactive TUI mode
     pinata agents chat <agent-id>

     # Single message with plain text response (for agents/scripts)
     echo "Hello" | pinata agents chat <agent-id> --text

     # JSONL output (machine-readable, default when piped)
     echo "Hello" | pinata agents chat <agent-id>

     # Multi-turn conversation (each line is a message)
     echo -e "Hello\nHow are you?" | pinata agents chat <id> -C --text

     # Interactive conversation from a file
     pinata agents chat <id> --conversation --text < messages.txt

     # Filter JSONL with jq
     echo "hi" | pinata agents chat <id> | jq -c 'select(.type=="content_delta")'

OPTIONS:
   --model string      Model override
   --json              Force JSONL output (auto-enabled when stdout is not a TTY)
   --text              Force plain text output (simpler alternative to JSONL for pipes)
   --conversation, -C  Multi-turn conversation mode (read messages from stdin line-by-line)
   --session string    Session key for conversation context (default: agent:main:cli)
   --help, -h          show help

Output Modes:

Mode	When	Use Case
Interactive TUI	TTY stdout	Human users in terminal
JSONL streaming	Non-TTY stdout (default)	Machine parsing, other agents
Plain text (`--text`)	Explicit flag	Simple scripts, easy piping
Multi-turn (`-C`)	Explicit flag	Conversations with context

Examples:

# Interactive TUI mode
pinata agents chat <agent-id>

# Single message with plain text response
echo "Hello" | pinata agents chat --text <agent-id>

# Multi-turn conversation (each line is a message)
echo -e "Hello\nWhat is 2+2?" | pinata agents chat --text -C <agent-id>

# JSONL output for machine parsing
echo "Hello" | pinata agents chat <agent-id>

# Filter JSONL to extract just the text
echo "Hello" | pinata agents chat <agent-id> | jq -r 'select(.type=="content_delta") | .delta.text' | tr -d '\n'

`exec`

Execute a command inside an agent’s container.

NAME:
   pinata agents exec - Execute a command in an agent container

USAGE:
   pinata agents exec [options] [agent ID] [command]

OPTIONS:
   --cwd string  Working directory for the command
   --help, -h    show help

Examples:

# Run a simple command
pinata agents exec <agent-id> 'echo hello'

# List files in workspace
pinata agents exec <agent-id> 'ls -la /app'

# Run with specific working directory
pinata agents exec <agent-id> --cwd /app 'cat config.json'

`files`

Read files from an agent’s container.

NAME:
   pinata agents files - Agent file operations

USAGE:
   pinata agents files [command [command options]]

COMMANDS:
   read, r  Read a file from agent container

OPTIONS:
   --help, -h  show help

Examples:

# Read a file from the agent
pinata agents files read <agent-id> /app/config.json

`devices`

Manage device pairing for agents (used for mobile/external device connections).

NAME:
   pinata agents devices - Manage agent devices

USAGE:
   pinata agents devices [command [command options]]

COMMANDS:
   list, l      List pending and paired devices
   approve, a   Approve a device pairing request
   approve-all  Approve all pending device requests

OPTIONS:
   --help, -h  show help

Examples:

# List devices for an agent
pinata agents devices list <agent-id>

# Approve a specific device request
pinata agents devices approve <agent-id> <request-id>

# Approve all pending requests
pinata agents devices approve-all <agent-id>

`skills`

NAME:
   pinata agents skills - Manage agent skills

USAGE:
   pinata agents skills [command [command options]]

COMMANDS:
   list, l    List available skills in library
   create, c  Create a new skill
   delete, d  Delete a skill from library
   attach, a  Attach skills to an agent
   detach     Detach a skill from an agent

OPTIONS:
   --help, -h  show help

create options:

OPTIONS:
   --cid string                     Content ID of the skill (required)
   --name string, -n string         Skill name (required)
   --description string, -d string  Skill description
   --env string [ --env string ]    Required environment variable names
   --file-id string                 Pinata v3 file ID
   --help, -h                       show help

Examples:

# List skills in your library
pinata agents skills list

# Create a skill from a CID
pinata agents skills create --cid <skill-cid> --name "My Skill" --env API_KEY

# Attach skills to an agent
pinata agents skills attach <agent-id> <skill-cid> <skill-cid>

# Detach a skill from an agent
pinata agents skills detach <agent-id> <skill-id>

`secrets`

NAME:
   pinata agents secrets - Manage secrets

USAGE:
   pinata agents secrets [command [command options]]

COMMANDS:
   list, l    List all secrets
   create, c  Create a new secret
   update, u  Update a secret value
   delete, d  Delete a secret
   attach, a  Attach secrets to an agent
   detach     Detach a secret from an agent

OPTIONS:
   --help, -h  show help

create options:

OPTIONS:
   --name string, -n string   Secret name (e.g. ANTHROPIC_API_KEY)
   --value string, -v string  Secret value
   --help, -h                 show help

Examples:

# Create a secret
pinata agents secrets create --name ANTHROPIC_API_KEY --value sk-ant-...

# Update a secret's value
pinata agents secrets update <secret-id> --value sk-ant-...

# Attach secrets to an agent
pinata agents secrets attach <agent-id> <secret-id> <secret-id>

`channels`

NAME:
   pinata agents channels - Manage agent channels

USAGE:
   pinata agents channels [command [command options]]

COMMANDS:
   status, s     Get channel configuration status
   configure, c  Configure a channel (telegram, slack, discord, whatsapp)
   remove, r     Remove a channel configuration

OPTIONS:
   --help, -h  show help

configure options:

OPTIONS:
   --bot-token string                           Bot token
   --app-token string                           App token (Slack only)
   --dm-policy string                           DM policy: open or pairing
   --allow-from string [ --allow-from string ]  Allowed user IDs/phone numbers
   --enabled                                    Enable or disable the channel
   --skip-restart                               Skip restarting the agent after configuring
   --help, -h                                   show help

Pass --skip-restart to batch multiple channel changes and avoid a gateway restart per call; the new configuration takes effect on the next restart. Examples:

# Get channel status
pinata agents channels status <agent-id>

# Configure Telegram
pinata agents channels configure <agent-id> telegram --bot-token <token>

# Configure a channel without restarting the agent
pinata agents channels configure <agent-id> telegram --bot-token <token> --skip-restart

# Disable a channel without removing its configuration
pinata agents channels configure <agent-id> telegram --enabled=false

# Remove a channel
pinata agents channels remove <agent-id> telegram

`snapshots`

NAME:
   pinata agents snapshots - Manage agent snapshots

USAGE:
   pinata agents snapshots [command [command options]]

COMMANDS:
   list, l    List agent snapshots
   create, c  Create a snapshot
   status, s  Get sync status
   reset, r   Reset to a snapshot

OPTIONS:
   --help, -h  show help

Examples:

# List snapshots
pinata agents snapshots list <agent-id>

# Check sync status
pinata agents snapshots status <agent-id>

# Create a snapshot
pinata agents snapshots create <agent-id>

# Reset to a snapshot
pinata agents snapshots reset <agent-id> <snapshot-cid>

`tasks`

NAME:
   pinata agents tasks - Manage agent cron jobs/tasks

USAGE:
   pinata agents tasks [command [command options]]

COMMANDS:
   list, l    List tasks
   create, c  Create a new task
   update, u  Update an existing task
   delete, d  Delete a task
   toggle     Enable or disable a task
   run        Run a task immediately
   history    View task run history

OPTIONS:
   --help, -h  show help

`create`

NAME:
   pinata agents tasks create - Create a new task

USAGE:
   pinata agents tasks create [options] [agent ID]

DESCRIPTION:
   Create a new cron job for an agent.

   Schedule types:
     --at       Run once at a specific time (ISO 8601 format)
     --every    Run at intervals (e.g., "1h", "30m", "24h")
     --cron     Run on a cron schedule (e.g., "0 9 * * *")

   Payload types (choose one):
     --system-event  System event text (triggers heartbeat-style execution)
     --agent-turn    Agent turn message (triggers conversational response)

   Examples:
     # Run every hour with a system event
     pinata agents tasks create <agent-id> --name "hourly-check" --every 1h --system-event "Check for updates"

     # Run daily at 9am UTC with agent turn
     pinata agents tasks create <agent-id> --name "daily-report" --cron "0 9 * * *" --agent-turn "Generate daily report"

     # Run once at a specific time
     pinata agents tasks create <agent-id> --name "one-time" --at "2026-04-01T12:00:00Z" --system-event "Do task"

OPTIONS:
   --name string, -n string           Task name (required)
   --description string               Task description
   --disabled                         Create task in disabled state
   --at string                        Run once at this time (ISO 8601)
   --every string                     Run every interval (e.g., 1h, 30m)
   --cron string                      Cron expression (e.g., '0 9 * * *')
   --tz string                        Timezone for cron schedule
   --system-event string              System event payload text
   --agent-turn string                Agent turn message
   --model string                     Model override for agent turn
   --timeout int                      Timeout in seconds (default: 0)
   --session string                   Session target: main or isolated (default: "main")
   --skill string [ --skill string ]  Skill CIDs to make available to the task (can be specified multiple times)
   --help, -h                         show help

`update`

NAME:
   pinata agents tasks update - Update an existing task

USAGE:
   pinata agents tasks update [options] [agent ID] [job ID]

DESCRIPTION:
   Update an existing cron job. Only specified fields are changed.

   Examples:
     # Change task name
     pinata agents tasks update <agent-id> <job-id> --name "new-name"

     # Update schedule to run every 2 hours
     pinata agents tasks update <agent-id> <job-id> --every 2h

     # Update payload message
     pinata agents tasks update <agent-id> <job-id> --system-event "New message"

OPTIONS:
   --name string, -n string           New task name
   --description string               New task description
   --at string                        Run once at this time (ISO 8601)
   --every string                     Run every interval (e.g., 1h, 30m)
   --cron string                      Cron expression (e.g., '0 9 * * *')
   --tz string                        Timezone for cron schedule
   --system-event string              System event payload text
   --agent-turn string                Agent turn message
   --model string                     Model override for agent turn
   --timeout int                      Timeout in seconds (default: 0)
   --skill string [ --skill string ]  Skill CIDs to make available to the task (can be specified multiple times)
   --help, -h                         show help

Scoping skills to a task: create and update accept --skill (repeatable) to make specific skill CIDs available only while the task runs:

# Create a task with task-scoped skills
pinata agents tasks create --name "nightly-report" --cron "0 0 * * *" \
  --agent-turn "Summarize today" --skill <skill-cid> --skill <skill-cid> <agent-id>

# Add or change task-scoped skills on an existing task
pinata agents tasks update <agent-id> <task-id> --skill <skill-cid>

`templates`

Browse and manage agent templates. Includes browsing published templates and submitting your own from a GitHub/GitLab repository.

NAME:
   pinata agents templates - Browse and manage agent templates

USAGE:
   pinata agents templates [command [command options]]

COMMANDS:
   list, l      List available templates
   get, g       Get template details by slug
   mine         List templates you have submitted
   validate, v  Validate a git repo for template submission
   submit, s    Submit a new template from a git repo
   update, u    Update an existing template submission (re-pull from repo)
   delete, d    Archive a template submission
   branches     List branches for a git repository
   refs         List branches and tags (with the default branch) for a git repository
   search-refs  Search branches and tags by name for a git repository

OPTIONS:
   --help, -h  show help

Browsing templates:

# List all templates
pinata agents templates list

# List featured templates only
pinata agents templates list --featured

# Filter by category
pinata agents templates list --category ai

# Get template details
pinata agents templates get ipfs-expert

Submitting templates: Your git repo must contain a valid manifest.json, a README.md, and a workspace/ directory. See Creating Templates for the full specification. submit, update, and validate share the --ref and --path flags:

OPTIONS:
   --ref string, -r string, --branch string, -b string  Git ref to submit from (e.g. refs/heads/main, refs/tags/v1.0.0, or a bare branch name; default: main)
   --path string, -p string                             Subdirectory within the repo to use as the template root (for monorepos)
   --name string                                        Override the template name from manifest.json
   --slug string                                        Override the template slug from manifest.json (lowercase, hyphens only)
   --help, -h                                           show help

--branch/-b are kept as aliases for --ref. You can pass a full ref (refs/heads/main, refs/tags/v1.0.0) or a bare branch name. Use --path to point at a subdirectory when your template lives inside a monorepo. --name/--slug are available on submit and update only.

# Validate your repo before submitting
pinata agents templates validate https://github.com/user/my-agent-template

# Validate a specific ref (branch or tag)
pinata agents templates validate https://github.com/user/my-agent-template --ref refs/tags/v1.0.0

# Validate a template that lives in a subdirectory (monorepo)
pinata agents templates validate https://github.com/user/monorepo --path packages/my-agent

# Submit the template
pinata agents templates submit https://github.com/user/my-agent-template

# Submit from a specific branch with name/slug overrides
pinata agents templates submit https://github.com/user/my-agent-template \
  --ref develop --name "My Agent" --slug my-agent

Discovering refs:

# List branches for a repo
pinata agents templates branches https://github.com/user/my-agent-template

# List branches and tags (with the default branch)
pinata agents templates refs https://github.com/user/my-agent-template

# Search branches and tags by name
pinata agents templates search-refs https://github.com/user/my-agent-template v1

Managing your submissions:

# List your submitted templates
pinata agents templates mine

# Update an existing submission (re-pull from repo)
pinata agents templates update <template-id>

# Update with a new repo URL or ref
pinata agents templates update <template-id> --git-url https://github.com/user/new-repo --ref main

# Archive a submission
pinata agents templates delete <template-id>

`clawhub`

Browse and install skills from ClawHub, the community skills marketplace.

NAME:
   pinata agents clawhub - Browse and install skills from ClawHub

USAGE:
   pinata agents clawhub [command [command options]]

COMMANDS:
   list, l     Browse ClawHub skills
   get, g      Get ClawHub skill details by slug
   install, i  Install a ClawHub skill to your library

OPTIONS:
   --help, -h  show help

list options:

OPTIONS:
   --category string, -c string  Filter by category
   --sort string, -s string      Sort by: popular, newest, name
   --featured                    Show only featured skills
   --cursor string               Pagination cursor
   --help, -h                    show help

Examples:

# Browse available skills
pinata agents clawhub list

# Browse featured skills, newest first
pinata agents clawhub list --featured --sort newest

# Get skill details by slug
pinata agents clawhub get clawdhub

# Install a skill to your library
pinata agents clawhub install <hub-skill-id>

`domains`

Manage custom domains for your agents (beta feature).

NAME:
   pinata agents domains - Manage custom domains (beta)

USAGE:
   pinata agents domains [command [command options]]

COMMANDS:
   list, l    List custom domains
   add, a     Register a custom domain
   update, u  Update a custom domain
   delete, d  Remove a custom domain

OPTIONS:
   --help, -h  show help

add options:

OPTIONS:
   --subdomain string  Subdomain name (e.g., 'myapp' for myapp.apps.pinata.cloud)
   --domain string     Custom domain (e.g., 'api.example.com')
   --port int          Target container port
   --protected         Require authentication
   --help, -h          show help

update options:

OPTIONS:
   --subdomain string  New subdomain name
   --domain string     New custom domain
   --port int          New target port (default: 0)
   --protected         Enable authentication
   --no-protected      Disable authentication
   --help, -h          show help

Examples:

# List domains for an agent
pinata agents domains list <agent-id>

# Add a subdomain (myapp.apps.pinata.cloud)
pinata agents domains add <agent-id> --subdomain myapp --port 8080

# Add a custom domain
pinata agents domains add <agent-id> --domain api.example.com --port 3000 --protected

# Update domain port
pinata agents domains update <agent-id> <domain-id> --port 8081

# Disable authentication on a domain
pinata agents domains update <agent-id> <domain-id> --no-protected

# Remove domain
pinata agents domains delete <agent-id> <domain-id>

`ports`

Manage agent port forwarding rules.

NAME:
   pinata agents ports - Manage agent port forwarding

USAGE:
   pinata agents ports [command [command options]]

COMMANDS:
   list, l  List port forwarding rules
   set, s   Set port forwarding rules

OPTIONS:
   --help, -h  show help

Examples:

# List current port mappings
pinata agents ports list <agent-id>

# Set port forwarding (port:pathPrefix format, up to 10 rules)
pinata agents ports set <agent-id> 8080:/api 3000:/app

# Clear all port forwarding rules
pinata agents ports set <agent-id>

`config`

Manage agent openclaw configuration.

NAME:
   pinata agents config - Manage agent configuration

USAGE:
   pinata agents config [command [command options]]

COMMANDS:
   get, g       Get agent openclaw config
   set, s       Set agent openclaw config (JSON)
   validate, v  Validate agent openclaw config

OPTIONS:
   --help, -h  show help

Examples:

# Read current config
pinata agents config get <agent-id>

# Validate config
pinata agents config validate <agent-id>

# Set config (JSON string)
pinata agents config set <agent-id> '{"key": "value"}'

`update`

Manage openclaw version updates.

NAME:
   pinata agents update - Manage agent openclaw updates

USAGE:
   pinata agents update [command [command options]]

COMMANDS:
   check, c  Check for openclaw updates
   apply, a  Apply openclaw update

OPTIONS:
   --help, -h  show help

apply options:

OPTIONS:
   --tag string  Specific version or tag to install (e.g., 'latest', '0.3.0', 'beta')
   --help, -h    show help

Examples:

# Check for updates
pinata agents update check <agent-id>

# Apply latest update
pinata agents update apply <agent-id>

# Apply specific version
pinata agents update apply <agent-id> --tag 2026.3.8

`versions`

List available agent image versions.

NAME:
   pinata agents versions - List available agent versions

USAGE:
   pinata agents versions [options] [agent ID]

OPTIONS:
   --help, -h  show help

Example:

pinata agents versions <agent-id>

For agents on an engine that publishes versions per engine (such as Hermes), the output includes availableVersionsPerEngine (grouped by openclaw/hermes) alongside the agent’s current engine and version.

`engines`

List the agent engines enabled for the current deployment. Use this to discover which values are valid for agents create --engine.

NAME:
   pinata agents engines - List the agent engines enabled for this deployment

USAGE:
   pinata agents engines [options]

OPTIONS:
   --help, -h  show help

Example:

pinata agents engines

Returns the enabled engines, each with an id (the value you pass to --engine) and a human-readable label:

[
  { "id": "openclaw", "label": "OpenClaw" },
  { "id": "hermes", "label": "Hermes" }
]

`auth`

Authenticate with AI providers and store credentials as secrets for your agents.

NAME:
   pinata agents auth - Authenticate with a provider and store the credential as a secret

USAGE:
   pinata agents auth [options] [provider: anthropic, openai, openrouter, venice]

OPTIONS:
   --oauth     Use OAuth browser flow instead of API key (openai only)
   --help, -h  show help

Supported Providers:

Provider	Secret Name	Options
`anthropic`	`ANTHROPIC_API_KEY`	-
`openai`	`OPENAI_API_KEY` or `OPENAI_OAUTH_TOKEN`	`--oauth` for Codex OAuth flow
`openrouter`	`OPENROUTER_API_KEY`	-
`venice`	`VENICE_API_KEY`	-

Examples:

# Store Anthropic API key
pinata agents auth anthropic

# Store OpenAI API key
pinata agents auth openai

# Authenticate with OpenAI via OAuth (for Codex CLI)
pinata agents auth openai --oauth

# Store OpenRouter API key
pinata agents auth openrouter

# Store Venice API key
pinata agents auth venice

`feedback`

Submit feedback or a feature request from the terminal.

NAME:
   pinata agents feedback - Submit feedback or feature request

USAGE:
   pinata agents feedback [options] [message]

OPTIONS:
   --help, -h  show help

Example:

pinata agents feedback "It would be great to..."

Documentation Index

New to Agents?

​list

​create

​get

​delete

​restart

​logs

​chat

​exec

​files

​devices

​skills

​secrets

​channels

​snapshots

​tasks

​create

​update

​templates

​clawhub

​domains

​ports

​config

​update

​versions

​engines

​auth

​feedback

`list`

`create`

`get`

`delete`

`restart`

`logs`

`chat`

`exec`

`files`

`devices`

`skills`

`secrets`

`channels`

`snapshots`

`tasks`

`create`

`update`

`templates`

`clawhub`

`domains`

`ports`

`config`

`update`

`versions`

`engines`

`auth`

`feedback`