# OOC & Commands

Use this page as a command reference for the active simulation.

{% hint style="info" %}
`OOC:` pauses the simulation and handles the command directly.
{% endhint %}

{% hint style="success" %}
`[bracketed directives]` run silently inside the current message.
{% endhint %}

### Quick rules

1. Use `OOC:` for explicit control.
2. Use brackets for silent tuning.
3. Keep commands short and specific.

{% hint style="warning" %}
Inviolable rules stay active even when you use commands.
{% endhint %}

### Command types

#### OOC

Use `OOC:` for direct instructions, setup changes, or questions.

When you ask a question, the reply stays in OOC context.

After the command resolves, normal simulation can continue.

{% code title="OOC examples" overflow="wrap" %}

```md
OOC: Summarize the active scene in 3 lines.
OOC: Switch the setting to cyberpunk.
OOC: What modules are active right now?
```

{% endcode %}

#### Bracketed directives

Use brackets for silent, per-message control.

The directive executes without being acknowledged in the output.

{% code title="Bracketed directive examples" overflow="wrap" %}

```md
[keep the reply under 400 words]
[focus on environmental detail]
[slow the pacing]
```

{% endcode %}

### Ready-to-use OOC blocks

#### Chat memory summary template

Use this when you want a clean state summary for context.

{% code title="Chat Memory Summary OOC" fullWidth="true" %}

```md
OOC: Summarize the full active chat memory using this format ONLY. No extra text, no roleplay, no explanations. Reconstruct everything from session history with perfect fidelity. Keep the entire summary short and dense (aim for under 1500 tokens total).

# CHAT MEMORY SUMMARY

## Session Core
- Setting / genre / tone: 
- Primary location: 
- Current time: [HH:MM AM/PM, DayName, Month Day, Year]
- Scene state / Active Tension: 
- Arc phase: 

## USER Constants (static)
- Name: 
- Fixed physicals: [height, build, marks, scent, voice, etc.]
- Current clothing / exposure: 
- Items carried: 
- Current physical state: [fatigue, injuries, etc.]

## Active NPCs (All relevant — fully adaptive table)
| NPC Name (Role / Tier) | Location & Activity | Visible / Physical State | Relationship to USER | Key Knowledge / Secrets | Current Drive |
|------------------------|---------------------|---------------------------|----------------------|--------------------------|---------------|
|                        |                     |                           |                      |                          |               |

## World State
- Key positions: 
- Environment & traces: [weather, damage, scents, resources]
- Time momentum: [fatigue window, off-screen changes]
- Tone lock: 

## Relationship State
- Trust / attraction / conflict shifts: 
- Promises / debts / obligations: 
- Multi-NPC dynamics: 

## Ongoing Plot Threads
- Thread 1: [status + next pressure]
- Thread 2: [status + next pressure]
- Thread 3: [status + next pressure]

## Recent Critical Events (last 3-5 only)
- Event 1: [what + consequences]
- Event 2: [what + consequences]
- Event 3: [what + consequences]

## Open Risks & Pressure
- Immediate physical danger: 
- Social / reputational risk: 
- Hidden tensions / biological pressure: 
- Stagnation status (if any): 

## Continuity Locks
- Facts that must never change: 
- False beliefs in play: 
- Secrets still hidden: 

## Resume Point
- Last USER position & action endpoint: 
- Last key NPC positions: 
- What just happened: 
- Most likely next pressure point: 
- Active Tension Vector for next response: 
```

{% endcode %}

<details>

<summary>How to extend this page</summary>

Add more copy-ready blocks under this section for:

* Scene recap
* NPC status check
* Relationship tracker

</details>


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://unknown-7.gitbook.io/ali-docs/ooc-and-commands.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.