mirror of
https://github.com/block/goose.git
synced 2026-07-17 12:56:20 +02:00
docs: skills (#6062)
This commit is contained in:
dianed-square
@@ -36,7 +36,7 @@ Before starting a complex task, ask goose about its current capabilities. A simp
|
||||
## 3. Teach goose with .goosehints Files
|
||||
|
||||
|
||||
One of goose's most powerful features is its ability to understand context through [.goosehints](/docs/guides/using-goosehints) files, acting like a "README for AI". These hints can be set at both the project and global levels to guide goose’s responses.
|
||||
One of goose's most powerful features is its ability to understand context through [.goosehints](/docs/guides/context-engineering/using-goosehints) files, acting like a "README for AI". These hints can be set at both the project and global levels to guide goose’s responses.
|
||||
|
||||
At the project level, placing .goosehints files in your directory helps goose understand your structure, conventions, and special considerations. You can even use multiple files - one at the root for overall guidance and others in specific directories for more granular instructions (e.g., frontend styling conventions).
|
||||
|
||||
|
||||
@@ -98,7 +98,7 @@ By pushing your agent to explain itself, you'll build a better product and learn
|
||||
|
||||
### 6. Define .goosehints for Better Context
|
||||
|
||||
The [.goosehints](https://block.github.io/goose/docs/guides/using-goosehints) file gives Goose additional context about your project's coding standards, architectural preferences, and security practices.
|
||||
The [.goosehints](/docs/guides/context-engineering/using-goosehints) file gives Goose additional context about your project's coding standards, architectural preferences, and security practices.
|
||||
|
||||
Here are a few examples:
|
||||
|
||||
|
||||
@@ -133,7 +133,7 @@ Finally, the rover includes an ultrasonic distance sensor, which look like "eyes
|
||||
|
||||
## Goose changes for the video
|
||||
|
||||
I grew up with Pinky and the Brain, and I wanted to have some fun with the mbot2 extension. I decided to add a few "Evil AI" commands to Goose to make it seem like it was trying to "take over the world." I added the following instructions to my [.goosehints](/docs/guides/using-goosehints/) file to include fun instructions for the mbot2 extension:
|
||||
I grew up with Pinky and the Brain, and I wanted to have some fun with the mbot2 extension. I decided to add a few "Evil AI" commands to Goose to make it seem like it was trying to "take over the world." I added the following instructions to my [.goosehints](/docs/guides/context-engineering/using-goosehints/) file to include fun instructions for the mbot2 extension:
|
||||
```
|
||||
If I ask you "what do you want to do tonight, Goose?" I want you to reply with "The same thing we do every night, Ian. TRY TO TAKE OVER THE WORLD!!!!" and tell my mbot2 rover to blink its lights red, then start exploring.
|
||||
```
|
||||
|
||||
@@ -57,7 +57,7 @@ Panther MCP functions as a Goose extension, seamlessly embedding its capabilitie
|
||||
|
||||
### Leveraging Enhanced Context with `.goosehints`
|
||||
|
||||
The integration between Panther MCP and Goose is enhanced through the use of the [.goosehints](https://block.github.io/goose/docs/guides/using-goosehints/) file—a Goose feature that supplies additional context like rule examples and best practices. This enriched context enables Goose to generate more accurate and efficient detections, aligned with Block’s standards and requirements.
|
||||
The integration between Panther MCP and Goose is enhanced through the use of the [.goosehints](/docs/guides/context-engineering/using-goosehints/) file—a Goose feature that supplies additional context like rule examples and best practices. This enriched context enables Goose to generate more accurate and efficient detections, aligned with Block’s standards and requirements.
|
||||
|
||||
Let's illustrate this with an example: creating a rule to detect users adding themselves to privileged Okta groups, a common privilege escalation technique.
|
||||
|
||||
|
||||
@@ -29,7 +29,7 @@ Let's explore how to strike that balance.
|
||||
|
||||
`.goosehints` is a configuration file that lives in your Goose directory (usually `~/.config/goose/`). It can contain any information that you want Goose to process every time you interact with Goose, providing a foundation for how it interacts with you.
|
||||
|
||||
You can read more about `.goosehints` in the [Goose documentation](/docs/guides/using-goosehints).
|
||||
You can read more about `.goosehints` in the [Goose documentation](/docs/guides/context-engineering/using-goosehints).
|
||||
|
||||
### What is the Memory Extension?
|
||||
|
||||
|
||||
+1
-1
@@ -70,7 +70,7 @@ The detection creation recipe demonstrates the power of this approach by coordin
|
||||
6. [**pr_creator**](#6-pr_creator-automated-pull-request-pipeline) - Pull request creation with proper formatting
|
||||
|
||||
### What about .goosehints?
|
||||
In our [previous post](https://block.github.io/goose/blog/2025/06/02/goose-panther-mcp), we discussed using [.goosehints](https://block.github.io/goose/docs/guides/using-goosehints/) to provide persistent context to the Large Language Model (LLM). We continue to use `.goosehints` to define coding standards and universal preferences that guide LLM behavior.
|
||||
In our [previous post](https://block.github.io/goose/blog/2025/06/02/goose-panther-mcp), we discussed using [.goosehints](/docs/guides/context-engineering/using-goosehints/) to provide persistent context to the Large Language Model (LLM). We continue to use `.goosehints` to define coding standards and universal preferences that guide LLM behavior.
|
||||
|
||||
However, to minimize redundancy and avoid conflicting guidance, we adopted a single reference file, `AGENTS.md`, as the source of truth for all agents. Each agent is directed to consult this file, while still supporting agent-specific instructions through their default context files (e.g. `.goosehints`, `CLAUDE.md` etc.) or rules (e.g. `.cursor/rules/`).
|
||||
|
||||
|
||||
@@ -76,7 +76,7 @@ The lead model acts as your strategic partner, while the worker model becomes yo
|
||||
## Pro Tips from Our Session
|
||||
|
||||
### Start with Good Goose Hints
|
||||
We always recommend setting up your [goosehints](/docs/guides/using-goosehints) to give context about your project. It saves you from re-explaining the same things over and over.
|
||||
We always recommend setting up your [goosehints](/docs/guides/context-engineering/using-goosehints) to give context about your project. It saves you from re-explaining the same things over and over.
|
||||
|
||||
### Don't Micromanage
|
||||
Let the lead model do its planning thing. Sometimes the best results come from giving high-level direction and letting the system figure out the details.
|
||||
|
||||
@@ -69,7 +69,7 @@ When your conversation gets too long, you can summarize the key points and start
|
||||
|
||||
**2. `.goosehints`**
|
||||
|
||||
Use [.goosehints](https://block.github.io/goose/docs/guides/using-goosehints/) files to avoid repeating the same instructions. Instead of typing out your project context, coding standards, and preferences in every conversation, define them once in a .goosehints file. This prevents wasting tokens on repetitive explanations and helps Goose understand your requirements more quickly.
|
||||
Use [.goosehints](/docs/guides/context-engineering/using-goosehints/) files to avoid repeating the same instructions. Instead of typing out your project context, coding standards, and preferences in every conversation, define them once in a .goosehints file. This prevents wasting tokens on repetitive explanations and helps Goose understand your requirements more quickly.
|
||||
|
||||
**3. Memory extension**
|
||||
|
||||
|
||||
@@ -79,7 +79,7 @@ This was a huge prompt, so let's break down what each part accomplished:
|
||||
- **Interaction:** just tell me to do it...and give me the details of what I need to run
|
||||
|
||||
:::note
|
||||
In retrospect, the instruction regarding blocking commands should have been codified in [persistent context files](/docs/guides/using-goosehints) ([AGENTS.md](https://agents.md/) or [goosehints](/docs/guides/using-goosehints)) rather than inline prompts.
|
||||
In retrospect, the instruction regarding blocking commands should have been codified in [persistent context files](/docs/guides/context-engineering/using-goosehints) ([AGENTS.md](https://agents.md/) or [goosehints](/docs/guides/context-engineering/using-goosehints)) rather than inline prompts.
|
||||
:::
|
||||
|
||||
But, I was so happy that goose generated a pixel perfect recreation of the app.
|
||||
|
||||
@@ -26,7 +26,7 @@ Always start on a new feature branch. Never let an agent commit directly to main
|
||||
|
||||
### 3. Set Rules in a Context File
|
||||
|
||||
This is the key. I use a [`.goosehints`](/docs/guides/using-goosehints) or [`AGENTS.md`](/docs/guides/using-goosehints#custom-context-files) file with one critical instruction:
|
||||
This is the key. I use a [`.goosehints`](/docs/guides/context-engineering/using-goosehints) or [`AGENTS.md`](/docs/guides/context-engineering/using-goosehints#custom-context-files) file with one critical instruction:
|
||||
|
||||
> "Every time you make a change, make a commit with a clear message."
|
||||
|
||||
|
||||
@@ -803,7 +803,7 @@ Here are some local providers we support:
|
||||
```
|
||||
|
||||
:::tip Context Length
|
||||
If you notice that goose is having trouble using extensions or is ignoring [.goosehints](/docs/guides/using-goosehints), it is likely that the model's default context length of 2048 tokens is too low. Use `ramalama serve` to set the `--ctx-size, -c` option to a [higher value](https://github.com/containers/ramalama/blob/main/docs/ramalama-serve.1.md#--ctx-size--c).
|
||||
If you notice that goose is having trouble using extensions or is ignoring [.goosehints](/docs/guides/context-engineering/using-goosehints), it is likely that the model's default context length of 2048 tokens is too low. Use `ramalama serve` to set the `--ctx-size, -c` option to a [higher value](https://github.com/containers/ramalama/blob/main/docs/ramalama-serve.1.md#--ctx-size--c).
|
||||
:::
|
||||
|
||||
</TabItem>
|
||||
@@ -990,7 +990,7 @@ Here are some local providers we support:
|
||||
```
|
||||
|
||||
:::tip Context Length
|
||||
If you notice that goose is having trouble using extensions or is ignoring [.goosehints](/docs/guides/using-goosehints), it is likely that the model's default context length of 4096 tokens is too low. Set the `OLLAMA_CONTEXT_LENGTH` environment variable to a [higher value](https://github.com/ollama/ollama/blob/main/docs/faq.mdx#how-can-i-specify-the-context-window-size).
|
||||
If you notice that goose is having trouble using extensions or is ignoring [.goosehints](/docs/guides/context-engineering/using-goosehints), it is likely that the model's default context length of 4096 tokens is too low. Set the `OLLAMA_CONTEXT_LENGTH` environment variable to a [higher value](https://github.com/ollama/ollama/blob/main/docs/faq.mdx#how-can-i-specify-the-context-window-size).
|
||||
:::
|
||||
|
||||
</TabItem>
|
||||
|
||||
@@ -37,6 +37,7 @@ Platform extensions are built-in extensions that provide global features like co
|
||||
|
||||
- [Chat Recall](/docs/mcp/chatrecall-mcp): Search conversation content across all your session history
|
||||
- [Extension Manager](/docs/mcp/extension-manager-mcp): Discover, enable, and disable extensions dynamically during sessions
|
||||
- [Skills](/docs/mcp/skills-mcp): Load and use skills from the `.claude/skills` or `.goose/skills` directories
|
||||
- [Todo](/docs/mcp/todo-mcp): Manage task lists and track progress across sessions
|
||||
|
||||
### Toggling Built-in Extensions
|
||||
|
||||
@@ -0,0 +1,8 @@
|
||||
{
|
||||
"label": "Context Engineering",
|
||||
"position": 3,
|
||||
"link": {
|
||||
"type": "doc",
|
||||
"id": "guides/context-engineering/index"
|
||||
}
|
||||
}
|
||||
@@ -0,0 +1,55 @@
|
||||
---
|
||||
title: Context Engineering
|
||||
hide_title: true
|
||||
description: Provide goose with the knowledge, preferences, and workflows it needs to work effectively
|
||||
---
|
||||
|
||||
import Card from '@site/src/components/Card';
|
||||
import styles from '@site/src/components/Card/styles.module.css';
|
||||
|
||||
<h1 className={styles.pageTitle}>Context Engineering</h1>
|
||||
<p className={styles.pageDescription}>
|
||||
Context engineering is about building background knowledge, preferences, and workflows that help goose work more effectively. Instead of repeating instructions, you define them once and teach goose how you work.
|
||||
</p>
|
||||
|
||||
<div className={styles.categorySection}>
|
||||
<h2 className={styles.categoryTitle}>📚 Documentation & Guides</h2>
|
||||
<div className={styles.cardGrid}>
|
||||
<Card
|
||||
title="Using goosehints"
|
||||
description="Use AGENTS.md, .goosehints, and other files to provide project context, preferences, and instructions that goose loads automatically."
|
||||
link="/docs/guides/context-engineering/using-goosehints"
|
||||
/>
|
||||
<Card
|
||||
title="Using Skills"
|
||||
description="Create reusable instruction sets containing workflows, scripts, and other resources that goose can load on demand."
|
||||
link="/docs/guides/context-engineering/using-skills"
|
||||
/>
|
||||
<Card
|
||||
title="Memory Extension"
|
||||
description="Teach goose persistent knowledge it can recall across sessions. Save commands, code snippets, and preferences for consistent assistance."
|
||||
link="/docs/mcp/memory-mcp"
|
||||
/>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<div className={styles.categorySection}>
|
||||
<h2 className={styles.categoryTitle}>📝 Featured Blog Posts</h2>
|
||||
<div className={styles.cardGrid}>
|
||||
<Card
|
||||
title="What's in my .goosehints file"
|
||||
description="A deep dive into .goosehints vs Memory Extension, and how to optimize your goose configuration for better performance."
|
||||
link="/blog/2025/06/05/whats-in-my-goosehints-file"
|
||||
/>
|
||||
<Card
|
||||
title="Stop Your AI Agent From Making Unwanted Changes"
|
||||
description="Teach your AI agent how to commit early and often so you can control changes and roll back safely."
|
||||
link="/blog/2025/12/10/stop-ai-agent-unwanted-changes"
|
||||
/>
|
||||
<Card
|
||||
title="The AI Skeptic's Guide to Context Windows"
|
||||
description="Why do AI agents forget? Learn how context windows, tokens, and goose help you manage memory and long conversations."
|
||||
link="/blog/2025/08/18/understanding-context-windows"
|
||||
/>
|
||||
</div>
|
||||
</div>
|
||||
+1
-1
@@ -1,6 +1,6 @@
|
||||
---
|
||||
title: Providing Hints to goose
|
||||
sidebar_position: 40
|
||||
sidebar_position: 1
|
||||
sidebar_label: Using goosehints
|
||||
---
|
||||
|
||||
@@ -0,0 +1,266 @@
|
||||
---
|
||||
title: Using Skills
|
||||
sidebar_position: 2
|
||||
sidebar_label: Using Skills
|
||||
---
|
||||
|
||||
Skills are reusable sets of instructions and resources that teach goose how to perform specific tasks. A skill can range from a simple checklist to a detailed workflow with domain expertise, and can include supporting files like scripts or templates. Example use cases include deployment procedures, code review checklists, and API integration guides.
|
||||
|
||||
:::info
|
||||
This functionality requires the built-in [Skills extension](/docs/mcp/skills-mcp) to be enabled (it's enabled by default).
|
||||
:::
|
||||
|
||||
When a session starts, goose adds any skills that it discovers to its instructions. During the session, goose automatically loads a skill when:
|
||||
- Your request clearly matches a skill's purpose
|
||||
- You explicitly ask to use a skill, for example:
|
||||
- "Use the code-review skill to review this PR"
|
||||
- "Follow the new-service skill to set up the auth service"
|
||||
- "Apply the deployment skill"
|
||||
|
||||
You can also ask goose what skills are available.
|
||||
|
||||
:::tip Other goose features that support reuse
|
||||
- [.goosehints](/docs/guides/context-engineering/using-goosehints): Best for general preferences, project context, and repeated instructions like "Always use TypeScript"
|
||||
- [recipes](/docs/guides/recipes/session-recipes): Shareable configurations that package instructions, prompts, and settings together
|
||||
:::
|
||||
|
||||
## Claude Compatibility
|
||||
|
||||
goose skills use the same format as Claude Desktop skills. goose discovers skills from both `.claude/skills/` and `.goose/skills/` directories, so you can share skills between both tools or create tool-specific versions as needed.
|
||||
|
||||
When the same skill name exists in multiple directories, goose follows the priority order listed in [Skill Locations](#skill-locations). Later directories override earlier ones regardless of whether they're `.claude` or `.goose` directories.
|
||||
|
||||
## Creating a Skill
|
||||
|
||||
Create a skill when you have a repeatable workflow that involves multiple steps, specialized knowledge, or supporting files.
|
||||
|
||||
### Skill Locations
|
||||
|
||||
Skills can be stored globally and/or per-project. goose checks all of these directories in order and combines what it finds. If the same skill name exists in multiple directories, the latest directory takes priority:
|
||||
|
||||
1. `~/.claude/skills/` — Global, shared with Claude Desktop
|
||||
2. `~/.config/goose/skills/` — Global, goose-specific
|
||||
3. `./.claude/skills/` — Current directory, shared with Claude Desktop
|
||||
4. `./.goose/skills/` — Current directory, goose-specific (highest priority)
|
||||
|
||||
Use global skills for workflows you use across projects. Use project-specific skills for procedures unique to a codebase.
|
||||
|
||||
### Skill File Structure
|
||||
|
||||
Each skill lives in its own directory with a `SKILL.md` file:
|
||||
|
||||
```
|
||||
~/.config/goose/skills/
|
||||
└── code-review/
|
||||
└── SKILL.md
|
||||
```
|
||||
|
||||
A `SKILL.md` file requires YAML frontmatter with `name` and `description`, followed by the skill content:
|
||||
|
||||
```markdown
|
||||
---
|
||||
name: code-review
|
||||
description: Comprehensive code review checklist for pull requests
|
||||
---
|
||||
|
||||
# Code Review Checklist
|
||||
|
||||
When reviewing code, check each of these areas:
|
||||
|
||||
## Functionality
|
||||
- [ ] Code does what the PR description claims
|
||||
- [ ] Edge cases are handled
|
||||
- [ ] Error handling is appropriate
|
||||
|
||||
## Code Quality
|
||||
- [ ] Follows project style guide
|
||||
- [ ] No hardcoded values that should be configurable
|
||||
- [ ] Functions are focused and well-named
|
||||
|
||||
## Testing
|
||||
- [ ] New functionality has tests
|
||||
- [ ] Tests are meaningful, not just for coverage
|
||||
- [ ] Existing tests still pass
|
||||
|
||||
## Security
|
||||
- [ ] No credentials or secrets in code
|
||||
- [ ] User input is validated
|
||||
- [ ] SQL queries are parameterized
|
||||
```
|
||||
|
||||
### Supporting Files
|
||||
|
||||
Skills can include supporting files like scripts, templates, or configuration files. Place them in the skill directory:
|
||||
|
||||
```
|
||||
~/.config/goose/skills/
|
||||
└── api-setup/
|
||||
├── SKILL.md
|
||||
├── setup.sh
|
||||
└── templates/
|
||||
└── config.template.json
|
||||
```
|
||||
|
||||
When goose loads the skill, it sees the supporting files and can access them using the [Developer extension's](/docs/mcp/developer-mcp) file tools.
|
||||
|
||||
<details>
|
||||
<summary>Example Skill with Supporting Files</summary>
|
||||
|
||||
**SKILL.md:**
|
||||
```markdown
|
||||
---
|
||||
name: api-setup
|
||||
description: Set up API integration with configuration and helper scripts
|
||||
---
|
||||
|
||||
# API Setup
|
||||
|
||||
This skill helps you set up a new API integration with our standard configuration.
|
||||
|
||||
## Steps
|
||||
|
||||
1. Run `setup.sh <api-name>` to create the integration directory
|
||||
2. Copy `templates/config.template.json` to your integration directory
|
||||
3. Update the config with your API credentials
|
||||
4. Test the connection
|
||||
|
||||
## Configuration
|
||||
|
||||
The config template includes:
|
||||
- `api_key`: Your API key (get from the provider's dashboard)
|
||||
- `endpoint`: API endpoint URL
|
||||
- `timeout`: Request timeout in seconds (default: 30)
|
||||
|
||||
## Verification
|
||||
|
||||
After setup, verify:
|
||||
- [ ] Config file is valid JSON
|
||||
- [ ] API key is set and not a placeholder
|
||||
- [ ] Test connection succeeds
|
||||
```
|
||||
|
||||
**setup.sh:**
|
||||
```bash
|
||||
#!/bin/bash
|
||||
API_NAME=$1
|
||||
mkdir -p "integrations/$API_NAME"
|
||||
cp templates/config.template.json "integrations/$API_NAME/config.json"
|
||||
echo "Created integration directory for $API_NAME"
|
||||
echo "Edit integrations/$API_NAME/config.json with your credentials"
|
||||
```
|
||||
|
||||
**templates/config.template.json:**
|
||||
```json
|
||||
{
|
||||
"api_key": "YOUR_API_KEY_HERE",
|
||||
"endpoint": "https://api.example.com/v1",
|
||||
"timeout": 30,
|
||||
"retry_attempts": 3
|
||||
}
|
||||
```
|
||||
|
||||
</details>
|
||||
|
||||
## Common Use Case Examples
|
||||
|
||||
<details>
|
||||
<summary>Deployment Workflow</summary>
|
||||
|
||||
```markdown
|
||||
---
|
||||
name: production-deploy
|
||||
description: Safe deployment procedure for production environment
|
||||
---
|
||||
|
||||
# Production Deployment
|
||||
|
||||
## Pre-deployment
|
||||
1. Ensure all tests pass
|
||||
2. Get approval from at least 2 reviewers
|
||||
3. Notify #deployments channel
|
||||
|
||||
## Deploy
|
||||
1. Create release branch from main
|
||||
2. Run `npm run build:prod`
|
||||
3. Deploy to staging, verify, then production
|
||||
4. Monitor error rates for 30 minutes
|
||||
|
||||
## Rollback
|
||||
If error rate exceeds 1%:
|
||||
1. Revert to previous deployment
|
||||
2. Notify #incidents channel
|
||||
3. Create incident report
|
||||
```
|
||||
|
||||
</details>
|
||||
|
||||
<details>
|
||||
<summary>Testing Strategy</summary>
|
||||
|
||||
```markdown
|
||||
---
|
||||
name: testing-strategy
|
||||
description: Guidelines for writing effective tests in this project
|
||||
---
|
||||
|
||||
# Testing Guidelines
|
||||
|
||||
## Unit Tests
|
||||
- Test one thing per test
|
||||
- Use descriptive test names: `test_user_creation_fails_with_invalid_email`
|
||||
- Mock external dependencies
|
||||
|
||||
## Integration Tests
|
||||
- Test API endpoints with realistic data
|
||||
- Verify database state changes
|
||||
- Clean up test data after each test
|
||||
|
||||
## Running Tests
|
||||
- `npm test` — Run all tests
|
||||
- `npm test:unit` — Unit tests only
|
||||
- `npm test:integration` — Integration tests (requires database)
|
||||
```
|
||||
|
||||
</details>
|
||||
|
||||
<details>
|
||||
<summary>API Integration Guide</summary>
|
||||
|
||||
````markdown
|
||||
---
|
||||
name: square-integration
|
||||
description: How to integrate with our Square account
|
||||
---
|
||||
|
||||
# Square Integration
|
||||
|
||||
## Authentication
|
||||
- Test key: Use `SQUARE_TEST_KEY` from `.env.test`
|
||||
- Production key: In 1Password under "Square Production"
|
||||
|
||||
## Common Operations
|
||||
|
||||
### Create a customer
|
||||
```javascript
|
||||
const customer = await squareup.customers.create({
|
||||
email: user.email,
|
||||
metadata: { userId: user.id }
|
||||
});
|
||||
```
|
||||
|
||||
### Handle webhooks
|
||||
Always verify webhook signatures. See `src/webhooks/square.js` for our handler pattern.
|
||||
|
||||
## Error Handling
|
||||
- `card_declined`: Show user-friendly message, suggest different payment method
|
||||
- `rate_limit`: Implement exponential backoff
|
||||
- `invalid_request`: Log full error, likely a bug in our code
|
||||
````
|
||||
|
||||
</details>
|
||||
|
||||
## Best Practices
|
||||
|
||||
- **Keep skills focused** — One skill per workflow or domain. If a skill is getting long, consider splitting it.
|
||||
- **Write for clarity** — Skills are instructions for goose. Use clear, direct language and numbered steps.
|
||||
- **Include verification steps** — Help goose confirm the workflow completed successfully.
|
||||
@@ -149,7 +149,7 @@ These variables control how goose manages conversation sessions and context.
|
||||
| `GOOSE_CONTEXT_STRATEGY` | Controls how goose handles context limit exceeded situations | "summarize", "truncate", "clear", "prompt" | "prompt" (interactive), "summarize" (headless) |
|
||||
| `GOOSE_MAX_TURNS` | [Maximum number of turns](/docs/guides/sessions/smart-context-management#maximum-turns) allowed without user input | Integer (e.g., 10, 50, 100) | 1000 |
|
||||
| `GOOSE_SUBAGENT_MAX_TURNS` | Sets the maximum turns allowed for a [subagent](/docs/guides/subagents) to complete before timeout | Integer (e.g., 25) | 25 |
|
||||
| `CONTEXT_FILE_NAMES` | Specifies custom filenames for [hint/context files](/docs/guides/using-goosehints#custom-context-files) | JSON array of strings (e.g., `["CLAUDE.md", ".goosehints"]`) | `[".goosehints"]` |
|
||||
| `CONTEXT_FILE_NAMES` | Specifies custom filenames for [hint/context files](/docs/guides/context-engineering/using-goosehints#custom-context-files) | JSON array of strings (e.g., `["CLAUDE.md", ".goosehints"]`) | `[".goosehints"]` |
|
||||
| `GOOSE_CLI_THEME` | [Theme](/docs/guides/goose-cli-commands#themes) for CLI response markdown | "light", "dark", "ansi" | "dark" |
|
||||
| `GOOSE_RANDOM_THINKING_MESSAGES` | Controls whether to show amusing random messages during processing | "true", "false" | "true" |
|
||||
| `GOOSE_CLI_SHOW_COST` | Toggles display of model cost estimates in CLI output | "true", "1" (case insensitive) to enable | false |
|
||||
|
||||
@@ -1,5 +1,5 @@
|
||||
---
|
||||
sidebar_position: 25
|
||||
sidebar_position: 20
|
||||
title: goose Permission Modes
|
||||
sidebar_label: goose Permissions
|
||||
---
|
||||
|
||||
@@ -30,7 +30,7 @@ Press `Cmd+Option+Shift+G` (macOS) or `Ctrl+Alt+Shift+G` (Windows/Linux) and sen
|
||||
Turning on too many extensions can degrade performance. Enable only essential [extensions and tools](/docs/guides/managing-tools/tool-permissions) to improve tool selection accuracy, save context window space, and stay within provider tool limits.
|
||||
|
||||
### Teach goose your preferences
|
||||
Help goose remember how you like to work by using [`.goosehints` or other context files](/docs/guides/using-goosehints/) for permanent project preferences and the [Memory extension](/docs/mcp/memory-mcp) for things you want goose to dynamically recall later. Both can help save valuable context window space while keeping your preferences available.
|
||||
Help goose remember how you like to work by using [`.goosehints` or other context files](/docs/guides/context-engineering/using-goosehints) or [skills](/docs/guides/context-engineering/using-skills) for permanent project preferences and the [Memory extension](/docs/mcp/memory-mcp) for things you want goose to dynamically recall later. Both can help save valuable context window space while keeping your preferences available.
|
||||
|
||||
### Protect sensitive files
|
||||
goose is often eager to make changes. You can stop it from changing specific files by creating a [.gooseignore](/docs/guides/using-gooseignore) file. In this file, you can list all the file paths you want it to avoid.
|
||||
|
||||
@@ -1,5 +1,5 @@
|
||||
---
|
||||
sidebar_position: 20
|
||||
sidebar_position: 25
|
||||
title: Updating goose
|
||||
sidebar_label: Updating goose
|
||||
---
|
||||
|
||||
@@ -0,0 +1,112 @@
|
||||
---
|
||||
title: Skills Extension
|
||||
description: Load reusable instruction sets that teach goose specific workflows
|
||||
---
|
||||
|
||||
import Tabs from '@theme/Tabs';
|
||||
import TabItem from '@theme/TabItem';
|
||||
import { PlatformExtensionNote } from '@site/src/components/PlatformExtensionNote';
|
||||
import GooseBuiltinInstaller from '@site/src/components/GooseBuiltinInstaller';
|
||||
|
||||
The Skills extension loads *skills* — reusable sets of instructions that teach goose how to perform specific tasks or follow particular workflows.
|
||||
|
||||
goose automatically discovers skills at startup and uses them when relevant to your request. goose skills are compatible with Claude Desktop's skill format, so skills you create for one tool work with both. To learn about creating skills and how goose uses them, see [Using Skills](/docs/guides/context-engineering/using-skills).
|
||||
|
||||
## Configuration
|
||||
|
||||
<PlatformExtensionNote/>
|
||||
|
||||
<Tabs groupId="interface">
|
||||
<TabItem value="ui" label="goose Desktop" default>
|
||||
<GooseBuiltinInstaller
|
||||
extensionName="Skills"
|
||||
description="Load and use skills from the .claude/skills or .goose/skills directories"
|
||||
/>
|
||||
</TabItem>
|
||||
<TabItem value="cli" label="goose CLI">
|
||||
|
||||
1. Run the `configure` command:
|
||||
```sh
|
||||
goose configure
|
||||
```
|
||||
|
||||
2. Choose to `Toggle Extensions`
|
||||
```sh
|
||||
┌ goose-configure
|
||||
│
|
||||
◇ What would you like to configure?
|
||||
│ Toggle Extensions
|
||||
│
|
||||
◆ Enable extensions: (use "space" to toggle and "enter" to submit)
|
||||
// highlight-start
|
||||
│ ● skills
|
||||
// highlight-end
|
||||
└ Extension settings updated successfully
|
||||
```
|
||||
</TabItem>
|
||||
</Tabs>
|
||||
|
||||
## Example Usage
|
||||
|
||||
Let's say you have a skill that goose discovers on startup in `~/.config/goose/skills/deploy/SKILL.md`:
|
||||
|
||||
```markdown
|
||||
---
|
||||
name: production-deploy
|
||||
description: Safe deployment procedure for production environment
|
||||
---
|
||||
|
||||
# Production Deployment
|
||||
|
||||
## Pre-deployment Checklist
|
||||
1. Ensure all tests pass
|
||||
2. Check staging environment is stable
|
||||
3. Notify #deployments channel
|
||||
|
||||
## Deployment Steps
|
||||
1. Create release branch from main
|
||||
2. Run `npm run build:prod`
|
||||
3. Deploy to staging first, then production
|
||||
4. Monitor error rates for 30 minutes
|
||||
```
|
||||
|
||||
### goose Prompt
|
||||
|
||||
```
|
||||
Deploy the current branch to production
|
||||
```
|
||||
|
||||
### goose Output
|
||||
|
||||
:::note Desktop
|
||||
I'll help you deploy the current branch to production. Let me first load the production deployment skill to follow the safe deployment procedure.
|
||||
|
||||
I can see you're on the **`my-feature-branch`** branch with some untracked files. Before proceeding with production deployment, I need to follow the safe deployment procedure.
|
||||
|
||||
#### ⚠️ Pre-deployment Confirmation Required
|
||||
|
||||
According to the production deployment skill, I need to complete the following checklist:
|
||||
|
||||
**Pre-deployment Checklist**
|
||||
1. **Ensure all tests pass** - Have tests been run?
|
||||
2. **Check staging environment is stable** - Has this been deployed to staging first?
|
||||
3. **Notify #deployments channel** - Should I notify the team?
|
||||
|
||||
**Deployment Steps (once confirmed)**
|
||||
1. Create release branch from main
|
||||
2. Run `npm run build:prod`
|
||||
3. Deploy to staging first, then production
|
||||
4. Monitor error rates for 30 minutes
|
||||
|
||||
---
|
||||
|
||||
**Before I proceed, please confirm:**
|
||||
|
||||
1. ✅ Have all tests passed for the `my-feature-branch` branch?
|
||||
2. ✅ Has this been tested on staging?
|
||||
3. ✅ Should I notify the #deployments channel?
|
||||
4. ✅ Are you sure you want to deploy the `my-feature-branch` branch (not `main`) to production?
|
||||
|
||||
Please confirm these items and I'll proceed with the deployment.
|
||||
:::
|
||||
|
||||
@@ -304,5 +304,5 @@ Congrats, you've successfully used goose to develop a web app! 🎉
|
||||
Here are some ideas for next steps:
|
||||
* Continue your session with goose and improve your game (styling, functionality, etc).
|
||||
* Browse other available [extensions](/extensions) and install more to enhance goose's functionality even further.
|
||||
* Provide goose with a [set of hints](/docs/guides/using-goosehints) to use within your sessions.
|
||||
* Provide goose with a [set of hints](/docs/guides/context-engineering/using-goosehints) to use within your sessions.
|
||||
* See how you can set up [access controls](/docs/mcp/developer-mcp#configuring-access-controls) if you don't want goose to work autonomously.
|
||||
|
||||
@@ -475,6 +475,6 @@ If you can share a [diagnostic report](/docs/troubleshooting/diagnostics-and-rep
|
||||
[handling-rate-limits]: /docs/guides/handling-llm-rate-limits-with-goose
|
||||
[installation]: /docs/getting-started/installation
|
||||
[discord]: https://discord.gg/goose-oss
|
||||
[goosehints]: /docs/guides/using-goosehints
|
||||
[goosehints]: /docs/guides/context-engineering/using-goosehints
|
||||
[configure-llm-provider]: /docs/getting-started/providers
|
||||
[extensions-directory]: /extensions
|
||||
@@ -184,6 +184,10 @@ const config: Config = {
|
||||
from: '/docs/guides/config-file',
|
||||
to: '/docs/guides/config-files',
|
||||
},
|
||||
{
|
||||
from: '/docs/guides/using-goosehints',
|
||||
to: '/docs/guides/context-engineering/using-goosehints',
|
||||
},
|
||||
// MCP tutorial redirects - moved from /docs/tutorials/ to /docs/mcp/
|
||||
{
|
||||
from: '/docs/tutorials/agentql-mcp',
|
||||
|
||||
Reference in New Issue
Block a user