> For the complete documentation index, see [llms.txt](https://helpdocs.conducttr.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://helpdocs.conducttr.com/labs/free-apps/colibri/colibri-skill-for-claude.md).

# Colibri Skill for Claude

The **Colibri skill** is a custom capability for Claude that turns a short description of a story, training scenario, or dilemma into a fully working **Conducttr Colibri branching exercise**.

Instead of building an interactive scenario by hand in Conducttr step by step, you describe what you want in plain language, and Claude generates the complete file for you.

**What makes this different from just asking Claude to "write a branching story":** Conducttr storytellers have taught Claude how to structure the player experience and how to generate a Colibri-compatible file ready to upload.

{% hint style="info" %}
Ask your account team for the Claude Skill.
{% endhint %}

### Installing the skill

1. Go to `Settings` in Claude.ai (click your profile icon, then `Settings`).
2. Navigate to `Customize` > `Skills` (on some plans this may appear under `Settings` > `Features`).
3. Click `+` to add a custom skill, then choose to upload a file.
4. Upload the skill file you were given (a `.skill` file).
5. Once uploaded, it appears in your Skills list. Make sure it's **toggled on**.

{% hint style="warning" %}
Custom skills are private to your own Claude account by default. Installing it doesn't share it with anyone else on your team unless you take a separate sharing action. 

If you ever need to update to a newer version of the skill, delete the old one from your Skills list first, then upload the replacement file the same way.
{% endhint %}

### How to use it

Once installed and toggled on, you don't need to do anything special to "activate" it, just describe what you want in a normal message, and Claude will recognise when the Colibri skill applies and use it automatically.

**Example prompts:**

> "Write a Colibri exercise where a salesperson is offered a subtle bribe by a prospective client."

> "Create a branching training scenario about a marketing team wanting to advertise 'unlimited' service despite a hidden usage cap."

> "Generate a Colibri story about an engineer who discovers a security vulnerability three days before a big client demo."

Claude will:

1. Ask a clarifying question only if your prompt is genuinely ambiguous (e.g. if it's unclear who the player actually is in the scenario), otherwise it will proceed and simply tell you what assumption it made
2. Design the story: the dilemma, the characters involved, the choices available at each step, and how those choices lead to different endings
3. Build the actual file and hand you a `.txps` file to download

You can then import that file directly into Conducttr.

<figure><img src="/files/dR9xRM1gXqZ0vyagHQax" alt=""><figcaption></figcaption></figure>

### What you'll get in the finished file

* A branching narrative with **real choices that matter.** Decisions meaningfully affect which ending is reached, not just cosmetic variation
* **Named characters** speaking in their own voice, cast to fit the story (for example, a manager applying pressure, an expert offering guidance, a colleague raising a concern)
* **Multiple distinct endings**, reflecting the different ways the scenario could play out based on the choices made
* Everything labelled and named appropriately within the file itself (including datapoints)

### What next?

Once loaded in the Editor you can update the style, replace our default character Archetypes with your personas, add rich media and change it up a notch if you want to lavish more attention on it.

{% hint style="info" %}
Consider using the **Colibri Visualiser** to see the branching. Find out more below:
{% endhint %}

{% content-ref url="/pages/A8UVjIkOiN1iI8F4HChX" %}
[Colibri Visualiser](/labs/free-apps/colibri/colibri-visualiser.md)
{% endcontent-ref %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## 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, and the optional `goal` query parameter:

```
GET https://helpdocs.conducttr.com/labs/free-apps/colibri/colibri-skill-for-claude.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.