> For the complete documentation index, see [llms.txt](https://notara.gitbook.io/notara-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://notara.gitbook.io/notara-docs/skills/using-skills.md).

# Using Skills

Skills can be invoked automatically by the agent, explicitly by users, or programmatically by external agents via MCP.

## Automatic Invocation

When you send a message to the agent, it analyzes the request and automatically activates any skills whose description closely matches your intent. You don't need to explicitly name a skill — if you ask something that sounds like "run a competitor analysis", and you have a `mkt-competitor-teardown` skill, the agent will activate it.

Automatic invocation works best when:

* Your skill descriptions are precise and unique (avoid generic names)
* The user's request clearly matches one skill's purpose
* No conflicting skills exist with similar descriptions

## Explicit Invocation

To invoke a skill by name, say:

```
@Notara use the eng-weekly-digest skill
```

Or with more context:

```
@Notara use the eng-error-digest skill for last night's errors in the API service
```

The skill slug is case-insensitive. You can also use the full skill name:

```
@Notara use the "Weekly Engineering Digest" skill
```

When a skill is explicitly invoked:

1. The skill's prompt is appended to the system context
2. The skill's bound tools become available
3. The skill's bound context is added
4. The agent processes your request with this enriched context

## Copying a Skill Prompt for External Use

Each skill has a **Copy Prompt** button on its card. This copies the full skill prompt to your clipboard, suitable for pasting into:

* Claude.ai directly
* ChatGPT
* Cursor's system prompt
* Any LLM interface

This is how you get value from your skills even outside the Notara agent — you built the prompt once, your team uses it everywhere.

## MCP: Skills as Prompts

When you connect an external agent to your Notara MCP endpoint, skills are exposed as **MCP prompts**. External agents can:

**List available skills:**

```
prompts/list
```

Returns all skills in the group, with their slugs and descriptions.

**Fetch a skill:**

```
prompts/get { name: "eng-error-digest" }
```

Returns the full prompt text, bound tool list, and any bundled asset paths.

**Example with Claude Code:**

```
@Notara what skills are available in this group?
```

Or in Claude Code's agent mode, you can explicitly request the skill via the MCP prompts interface.

### Skill Packages via MCP

If a skill has bundled assets (reference files, example outputs), they're accessible as MCP resources under the `notara://skill/{slug}/asset/{path}` URI scheme. External agents can fetch these resources to get the full context the skill was designed with.

## Skill Invocation in Automations

Skills are commonly used in scheduled automations — run a skill automatically every week without anyone needing to type a command.

Example: a scheduled task runs every Monday at 8am, invokes the `growth-weekly-digest` skill, and posts the result to `#growth`. The task prompt is:

```
Run the growth-weekly-digest skill and post the result.
```

See [Scheduled Tasks](/notara-docs/automations/scheduled.md) for how to set this up.

## When Multiple Skills Match

If multiple skills could match a request, the agent picks the most specific one. If it's ambiguous, it asks the user to clarify or lists the matching skills with a description of each.

You can prevent ambiguity by:

* Making skill descriptions distinct and specific
* Assigning skills to specific channels (so only relevant skills are active in each channel)
* Giving skills unique, unambiguous names

## Skill Readiness

A skill is "ready" when all its bound tools are connected and have valid credentials. On the skill shelf:

* **Green dot** — all tools connected
* **Amber dot** — one or more tools need connecting (the skill will still work, but those tools will be unavailable)
* **Grey** — uses platform-managed tools (Firecrawl, Perplexity) — always available

To fix an amber readiness state, go to **Tools** and connect the missing connector.


---

# 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://notara.gitbook.io/notara-docs/skills/using-skills.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.