> ## Documentation Index
> Fetch the complete documentation index at: https://comis-fix-skill-import-vetting-gate.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Slash Commands

> Verified chat commands for sessions, models, execution, and approvals

Comis recognizes commands only at the start of a message. Unknown commands pass
through as ordinary message text.

## Available commands

| Command                                                                  | Behavior                                                                      |
| ------------------------------------------------------------------------ | ----------------------------------------------------------------------------- |
| `/think [off\|minimal\|low\|medium\|high\|xhigh]`                        | Sets the reasoning level for the execution; without a level it uses `high`    |
| `/context`                                                               | Shows available bootstrap-file and tool-schema context information            |
| `/status`                                                                | Shows session, message, token, context, and budget information when available |
| `/usage`                                                                 | Shows recorded usage by provider and model                                    |
| `/config [show [section]\|set <section.key> <value>\|history]`           | Reads or changes runtime configuration; requires admin trust                  |
| `/model [status\|list\|cycle\|next\|prev\|previous\|<provider>/<model>]` | Inspects or changes the session model                                         |
| `/new [provider/model]`                                                  | Destroys the current conversation and creates a new session                   |
| `/reset`                                                                 | Clears the current conversation state                                         |
| `/compact [verbose\|-v] [instructions]`                                  | Requests context compaction                                                   |
| `/export [path]`                                                         | Requests an HTML session export                                               |
| `/export-trajectory`                                                     | Sends an owner-gated trajectory bundle for the current direct-message session |
| `/stop`                                                                  | Aborts the active execution, if one exists                                    |
| `/fork`                                                                  | Forks the live conversation at the latest user message                        |
| `/branch [target-id]`                                                    | With no target, lists branch points; with an ID, navigates to it              |
| `/budget <Nk\|Nm> [message]`                                             | Applies a token budget to an execution; accepted range is 10k–10m             |

Examples:

```text theme={}
/think high
/model openai/gpt-4o
/branch
/budget 500k Research this repository and summarize the risks.
```

<Note>
  The available thinking levels depend on the model SDK. A provider can ignore a
  level that its model does not support. Cost and context fields also display only
  when the runtime exposes those measurements.
</Note>

## Conditional control commands

These commands are handled before the general command dispatcher:

| Command                    | Availability                                                             |
| -------------------------- | ------------------------------------------------------------------------ |
| `/approve [short-id\|all]` | Resolves a pending approval when the approval gate is enabled and wired  |
| `/deny [short-id\|all]`    | Denies a pending approval under the same conditions                      |
| `/send on\|off\|inherit`   | Lets the session owner override the runtime send policy for that session |

Approvals are disabled by default. Enable and configure them before relying on
`/approve` or `/deny`; see [Approvals](/security/approvals).

## Prompt skills are not slash commands

The current daemon does not wire `/skill:name` to direct skill loading. A string
such as `/skill:weather` is ordinary text and may be interpreted by the model,
but it is not a deterministic command. Prompt skills are discovered and offered
to the agent for model-selected use. See [Prompt Skills](/skills/prompt-skills).

Likewise, `/verbose` and `/reasoning` are parsed internally but their directives
are not currently consumed by the agent execution path. They are intentionally
omitted from the supported command table.

## Scope and delivery

Commands are processed by Comis, not by a messaging platform's native slash
command system. Availability can still depend on daemon wiring, sender trust,
session ownership, and whether a live SDK session exists. A command that needs a
live execution surface reports that it is unavailable when the surface is absent.

Deterministic command, approval, and gate responses return to the complete
authenticated turn endpoint, including the configured channel instance,
conversation, and source thread. A response to a threaded command never falls
back to the channel root.

<CardGroup cols={2}>
  <Card title="Sessions" icon="comments" href="/agents/sessions">
    How conversations are stored and managed
  </Card>

  <Card title="Models" icon="microchip" href="/agents/models">
    Configure providers and model selection
  </Card>

  <Card title="Prompt Skills" icon="wand-magic-sparkles" href="/skills/prompt-skills">
    Create model-selected reusable instructions
  </Card>

  <Card title="Approvals" icon="clipboard-check" href="/security/approvals">
    Enable human approval for selected actions
  </Card>
</CardGroup>