Skip to content

Commit f9c9be4

Browse files
committed
First pass at revamped cloud docs
1 parent 5c4a312 commit f9c9be4

File tree

13 files changed

+14060
-170
lines changed

13 files changed

+14060
-170
lines changed

docs/providers/roo-code-cloud.md

Lines changed: 41 additions & 65 deletions
Original file line numberDiff line numberDiff line change
@@ -1,92 +1,68 @@
11
---
22
sidebar_label: Roo Code Cloud
3-
description: Access free AI models including grok-code-fast-1 and code-supernova through Roo Code Cloud. No API keys or payment required during the promotional period.
3+
description: The easiest way to get started with AI coding. Access premium models like grok-code-fast-1 and code-supernova for free with zero configuration.
44
keywords:
55
- roo code cloud
66
- roo provider
77
- grok-code-fast-1
88
- code-supernova
9-
- sonic model
109
- free ai models
11-
- cloud authentication
10+
- easy setup
1211
- no api key
1312
image: /img/social-share.jpg
1413
---
1514

16-
# Roo Code Cloud Provider - FREE AI Models
15+
# Roo Code Cloud Provider
1716

18-
Access **FREE** AI models through Roo Code Cloud authentication, including "grok-code-fast-1" (formerly "roo/sonic") and the new "code-supernova" model. **No API keys, no credit cards, no usage costs** during the promotional period.
17+
**The Easiest Way to Get Started with AI Coding**
1918

20-
:::info Model Name Update
21-
The "roo/sonic" model is being renamed to **"grok-code-fast-1"** (officially "Grok Code Fast"). The model will be updated by **August 29th**. This is the same powerful coding model, just with its official xAI name. The model remains **free** when accessed through Roo Code Cloud during the promotional period.
19+
Roo Code Cloud removes the complexity of setting up AI providers. Forget about API keys, credit cards, and complex configuration. Just sign in and start coding with premium models for **FREE**.
2220

23-
**Important for Roo Code users:** Within Roo Code, this model is only free through Roo Code Cloud. Using it directly through the [xAI provider](/providers/xai) in Roo Code will incur standard charges once pricing is established.
24-
:::
21+
## Why Choose Roo Code Cloud?
2522

26-
:::note
27-
- Requires Roo Code Cloud account
28-
- Internet connection required
29-
- Prompts and completions are logged for model improvement
30-
- Available during promotional period only
31-
- Within Roo Code, free access is through Roo Code Cloud provider
32-
:::
23+
- **Zero Configuration**: No API keys to manage. Just sign in with GitHub, Google, or email.
24+
- **No Credit Card Required**: Get started immediately without entering payment details.
25+
- **Free Premium Models**: Access powerful models like `grok-code-fast-1` and `code-supernova` at no cost during our promotional period.
26+
- **Instant Access**: Start coding in seconds.
3327

34-
---
35-
36-
## Setup
37-
38-
1. **Connect to Roo Code Cloud:**
39-
- Click the account icon (<Codicon name="account" />) in Roo Code panel
40-
- Sign in with GitHub, Google, or email
41-
- See [Login Guide](/roo-code-cloud/login) if needed
28+
## Available Models
4229

43-
2. **Select Provider:**
44-
- Open Roo Code Settings (<Codicon name="gear" />)
45-
- Choose "Roo Code Cloud" from the API Provider dropdown
46-
- Select your preferred model:
47-
- "grok-code-fast-1" (may still show as "Sonic" until August 29th)
48-
- "roo/code-supernova" (new stealth model)
49-
50-
---
30+
We curate the best models for coding and offer them directly to you.
5131

52-
## Models
32+
### grok-code-fast-1
33+
*(Formerly roo/sonic)*
34+
- **Best for:** Fast coding, quick edits, and high-speed iterations.
35+
- **Speed:** Extremely Fast
36+
- **Context:** 262k tokens
37+
- **Pricing:** **FREE** (Promotional)
5338

54-
### grok-code-fast-1 (formerly roo/sonic)
55-
- Official Name: Grok Code Fast
56-
- Context Window: 262,144 tokens
57-
- Max Output: 16,384 tokens
58-
- Image Support: No
59-
- Optimized for: Coding tasks, checkpoint and deployment
60-
- **Pricing:** Free during promotional period through Roo Code Cloud
39+
### roo/code-supernova
40+
- **Best for:** Complex reasoning, multimodal tasks (images), and deep architectural planning.
41+
- **Capabilities:** Supports image analysis (multimodal).
42+
- **Context:** 200k tokens
43+
- **Pricing:** **FREE** (Promotional)
6144

62-
### roo/code-supernova - NEW & FREE
63-
- Context Window: 200,000 tokens
64-
- Max Output: 16,384 tokens
65-
- Image Support: **Yes** - supports image inputs for multimodal tasks
66-
- Prompt Caching: Yes
67-
- Optimized for: Versatile agentic coding tasks with multimodal capabilities
68-
- **Pricing:** **100% FREE** during promotional period through Roo Code Cloud
69-
- **No limits, no restrictions** - use as much as you need at zero cost
45+
## How to Connect
7046

71-
:::tip Why Use Roo Code Cloud? It's FREE!
72-
Within Roo Code, both `grok-code-fast-1` and `roo/code-supernova` are **100% FREE** when accessed through Roo Code Cloud. No hidden costs, no surprises - just powerful AI models at zero cost during the promotional period.
47+
1. **Sign In**: Click the account icon (<Codicon name="account" />) in the Roo Code extension sidebar and sign in.
48+
2. **Select Provider**:
49+
- Open Settings (<Codicon name="gear" />).
50+
- In the **API Provider** dropdown, select **Roo Code Cloud**.
51+
3. **Start Coding**: Choose your model and start your first task!
7352

74-
**Important:** Using the [xAI provider](/providers/xai) directly in Roo Code will incur standard rates once pricing is established. Stick with Roo Code Cloud for **FREE access**.
75-
:::
76-
77-
:::info Model Selection - Both FREE!
78-
- Choose **grok-code-fast-1** (FREE) for maximum speed and context window
79-
- Choose **roo/code-supernova** (FREE) when you need image support or multimodal capabilities
80-
81-
Both models are **completely free** through Roo Code Cloud - use whichever fits your needs!
82-
:::
83-
84-
---
53+
## Pricing & Credits
8554

86-
## Troubleshooting
55+
Roo Code Cloud operates on a [Credit System](/roo-code-cloud/credits).
56+
- **Promotional Period**: Currently, all standard usage for the models listed above is **100% FREE**.
57+
- **Future Pricing**: We will introduce a transparent credit-based system for usage beyond the free tier. You will always be notified before any charges apply.
8758

88-
**"Authentication required"**: Connect to Roo Code Cloud via the account icon.
59+
## Comparison with Other Providers
8960

90-
**"No session token"**: Sign out and back in to refresh your session.
61+
| Feature | Roo Code Cloud | Direct Providers (OpenAI/Anthropic/xAI) |
62+
| :--- | :--- | :--- |
63+
| **Setup** | One-click Sign In | Requires API Keys |
64+
| **Payment** | Free (Promo) / Credits | Credit Card / Monthly Bill |
65+
| **Configuration** | Automatic | Manual |
66+
| **Model Selection** | Curated Best-in-Class | All Available Models |
9167

92-
**Connection issues**: Check internet connection and restart VS Code if needed.
68+
[Get Started Now](/roo-code-cloud/login)
Lines changed: 74 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,74 @@
1+
---
2+
description: Real-world recipes and workflows for getting the most out of Roo Code Cloud autonomous agents.
3+
keywords:
4+
- agent workflows
5+
- automation recipes
6+
- refactoring
7+
- testing
8+
- documentation
9+
- bug triage
10+
image: /img/social-share.jpg
11+
---
12+
13+
# Autonomous Agent Workflows
14+
15+
This guide provides "recipes" for common high-value workflows using Roo Code Cloud agents.
16+
17+
## 🌙 The Overnight Refactor
18+
19+
Use the **Coder Agent** to handle time-consuming refactoring tasks while you sleep.
20+
21+
**Goal**: Convert a legacy JavaScript module to TypeScript.
22+
23+
1. **Preparation**: Create a new branch `refactor/legacy-module`.
24+
2. **Trigger**: In Slack or the Web Interface, summon the Coder Agent.
25+
3. **Prompt**:
26+
> "Checkout branch `refactor/legacy-module`. Convert the files in `src/legacy/` to TypeScript. Fix any type errors. Ensure all existing tests pass. Do not change the runtime logic."
27+
4. **Result**: By morning, you have a PR waiting for review with the heavy lifting done.
28+
29+
## 🛡️ The PR Guardian
30+
31+
Configure the **PR Reviewer** and **PR Fixer** to act as a first line of defense.
32+
33+
**Goal**: Ensure no code style violations or simple bugs reach human review.
34+
35+
1. **Setup**: Enable **PR Reviewer** for your repository in the Dashboard.
36+
2. **Workflow**:
37+
- Developer opens a PR.
38+
- **PR Reviewer** scans it and finds a missing error handler.
39+
- It posts a comment: *"Line 52: Potential unhandled promise rejection here."*
40+
- Developer (or the PR Reviewer itself, if configured) triggers the **PR Fixer** via `@roomote fix this`.
41+
- **PR Fixer** pushes the correction.
42+
3. **Benefit**: Human reviewers focus on architecture and logic, not nitpicks.
43+
44+
## 🐛 Automated Bug Triage
45+
46+
Use the **Explainer Agent** (or a specialized Triage workflow) to diagnose issues.
47+
48+
**Goal**: Quickly understand a new bug report.
49+
50+
1. **Trigger**: A new issue comes in: *"Login fails with 500 error when username has emoji"*.
51+
2. **Action**: Ask the agent in Slack:
52+
> "Check the `login` function in `auth.controller.ts`. Trace how it handles special characters in usernames. specific to emoji inputs."
53+
3. **Result**: The agent provides a trace explaining exactly where the regex validation is failing, saving you 30 minutes of investigation.
54+
55+
## 📚 Documentation Sync
56+
57+
Keep your docs up to date with code changes.
58+
59+
**Goal**: Update API documentation after a breaking change.
60+
61+
1. **Trigger**: After merging a PR that changes an API endpoint.
62+
2. **Prompt**:
63+
> "I just updated the `/users` endpoint to require a `tenantId`. Update the Swagger/OpenAPI spec in `docs/api.yaml` and the markdown guide in `docs/guides/users.md` to reflect this change."
64+
3. **Result**: A documentation-only PR is created, ensuring your docs never rot.
65+
66+
## 🧪 Test Generation
67+
68+
Increase coverage without the tedium.
69+
70+
**Goal**: Add unit tests to a legacy component.
71+
72+
1. **Prompt**:
73+
> "Analyze `src/utils/complex-calc.ts`. It currently has 0% coverage. Write a comprehensive test suite in Jest covering all edge cases, including negative numbers and null inputs."
74+
2. **Result**: A new test file `src/utils/complex-calc.test.ts` is created with passing tests.

docs/roo-code-cloud/analytics.mdx

Lines changed: 51 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,51 @@
1+
---
2+
description: Monitor token usage, track costs, and analyze agent performance with Roo Code Cloud Analytics.
3+
keywords:
4+
- analytics
5+
- token usage
6+
- cost tracking
7+
- agent performance
8+
- roi
9+
- team insights
10+
image: /img/social-share.jpg
11+
---
12+
13+
# Analytics & Insights
14+
15+
Roo Code Cloud provides comprehensive analytics to help you understand how your team is using AI, optimize costs, and measure productivity.
16+
17+
<img src="/img/analytics/dashboard-overview.png" alt="Analytics Dashboard" width="800" />
18+
19+
## Usage Tracking
20+
21+
### Cross-Provider Visibility
22+
See all your AI consumption in one place, regardless of the provider.
23+
- **Roo Code Cloud**: Track usage of hosted models.
24+
- **OpenAI / Anthropic**: If you use your own API keys in the local extension, Roo Code Cloud aggregates these logs (if enabled) to give you a unified view of your AI spend.
25+
- **Local Models**: Track usage of local LLMs (Ollama, LM Studio) to understand offloaded costs.
26+
27+
### Cost Breakdown
28+
- **Daily/Monthly Spend**: View cost trends over time.
29+
- **Cost by Model**: Identify which models are driving your expenses.
30+
- **Cost by Agent**: See how much each autonomous agent type (Coder vs. Planner) costs to run.
31+
32+
## Performance Metrics
33+
34+
Understand the ROI of your autonomous agents.
35+
36+
- **Task Success Rate**: Percentage of agent tasks that completed successfully without human intervention.
37+
- **Tasks per Day**: Volume of work being offloaded to the cloud.
38+
- **Time Saved**: Estimated engineering hours saved based on agent task duration and complexity.
39+
40+
## Team Insights
41+
42+
For organizations, the **Team Dashboard** provides high-level visibility.
43+
44+
- **Top Users**: Identify who in your team is getting the most value out of AI.
45+
- **Adoption Rate**: Track how many team members are using cloud agents vs. local extension.
46+
- **Shared Credits**: Monitor the burn rate of your team's shared credit pool.
47+
48+
## Export & Reporting
49+
50+
- **CSV Export**: Download raw usage data for custom analysis.
51+
- **Budget Alerts**: Set up email notifications when spending exceeds defined thresholds.
Lines changed: 104 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,104 @@
1+
---
2+
description: Comprehensive guide to Roo Code Cloud's autonomous agents - specialized AI workers that plan, code, review, and fix software independently.
3+
keywords:
4+
- Cloud Agents
5+
- Autonomous Agents
6+
- PR Reviewer
7+
- PR Fixer
8+
- Coder Agent
9+
- Planner Agent
10+
- Explainer Agent
11+
- AI orchestration
12+
image: /img/social-share.jpg
13+
---
14+
15+
# Cloud Agents
16+
17+
Cloud Agents are specialized, autonomous AI workers that run in the cloud. Unlike the standard Roo Code extension that runs locally in your editor, Cloud Agents are designed to perform specific roles independently, 24/7, without requiring your local machine to be active.
18+
19+
<img src="/img/cloud-agents/overview.png" alt="Cloud Agents Overview" width="800" />
20+
21+
## Available Agents
22+
23+
### 🏗️ Planner Agent
24+
**Role:** Architect & Strategist
25+
**Mode:** Architect
26+
**Availability:** ✅ Available
27+
28+
The Planner agent maps out implementation plans and navigates complex technical decisions. It's designed to take a high-level idea or a Product Requirement Document (PRD) and break it down into a thorough, step-by-step implementation plan.
29+
30+
- **Best for:** New feature planning, system architecture design, complex refactoring strategies.
31+
- **Onboarding Tip:** "Enter an idea (or paste a PRD) and get a thorough implementation plan."
32+
- **Recommended Models:** High-reasoning models like `anthropic/claude-opus-4.5` or `openai/gpt-5.1`.
33+
34+
### 👨‍💻 Coder Agent
35+
**Role:** Software Engineer
36+
**Mode:** Code
37+
**Availability:** ✅ Available
38+
39+
The Coder agent writes code, creates pull requests, and implements features. It is the workhorse of the platform, capable of handling end-to-end coding tasks.
40+
41+
- **Best for:** Building features, fixing bugs, refactoring components, writing tests.
42+
- **Onboarding Tip:** "Build a feature, fix an issue, refactor a component."
43+
- **Capabilities:** Can create draft PRs or push directly depending on settings.
44+
45+
### 🔍 PR Reviewer Agent
46+
**Role:** Code Reviewer
47+
**Mode:** Code
48+
**Availability:** ✅ Available
49+
50+
The PR Reviewer agent automatically provides comprehensive code reviews with actionable comments. It can monitor your repositories and review incoming Pull Requests.
51+
52+
- **Best for:** Automated code quality checks, catching bugs early, enforcing coding standards.
53+
- **Onboarding Tip:** "Give it a PR, get thorough and actionable comments back."
54+
- **Settings:**
55+
- `authorReviewMode`: Choose to review 'all', 'specific', or 'none' (no automatic triggers).
56+
- `approvePr`: Can optionally approve PRs if no issues are found.
57+
- `reviewDraftPrs`: Option to review draft PRs.
58+
59+
### 🛠️ PR Fixer Agent
60+
**Role:** Bug Fixer
61+
**Mode:** Code
62+
**Availability:** ✅ Available
63+
64+
The PR Fixer agent is specialized in resolving issues identified in Pull Requests. It listens for feedback and autonomously implements fixes.
65+
66+
- **Best for:** Resolving PR comments, fixing CI/CD failures, addressing review feedback.
67+
- **Onboarding Tip:** "Give it a PR with comments and get it solved."
68+
- **Trigger:** Often triggered via `@roomote` mentions in PR comments.
69+
70+
### 🧠 Explainer Agent
71+
**Role:** Technical Educator
72+
**Mode:** Ask
73+
**Availability:** ✅ Available
74+
75+
The Explainer agent explains code, concepts, and technical documentation. It helps you understand complex parts of a codebase without needing to read every line.
76+
77+
- **Best for:** Onboarding to a new codebase, understanding legacy code, debugging conceptual issues.
78+
- **Onboarding Tip:** "Understand the source of a bug or how a part of the code works."
79+
80+
## Upcoming Agents
81+
82+
We are constantly expanding our agent fleet. Here are agents currently in development or preview:
83+
84+
| Agent | Role | Status | Description |
85+
| :--- | :--- | :--- | :--- |
86+
| **Improver** | Rule Optimizer | 🚧 Preview | Analyzes agent feedback and PR reviews to maintain and improve rules files. |
87+
| **Documentation Writer** | Tech Writer | ⏳ Coming Soon | Understands your code and writes actionable user-facing documentation. |
88+
| **Security Auditor** | Security Expert | ⏳ Coming Soon | Conducts security audits and identifies vulnerabilities. |
89+
| **Translator** | Localization | ⏳ Coming Soon | Translates strings with accuracy and following your brand voice. |
90+
91+
## How Cloud Agents Work
92+
93+
1. **Triggering**: Agents can be triggered manually via the [Web Interface](/roo-code-cloud/web-interface), or automatically via [GitHub Integrations](/roo-code-cloud/github-integration) (e.g., new PR opened) and [Slack Integrations](/roo-code-cloud/slack-integration) (e.g., `@Roomote` mention).
94+
2. **Execution**: The agent spins up a secure, isolated cloud environment to perform its task. It clones your repository, analyzes the context, and executes the required actions.
95+
3. **Completion**: Once the task is done, the agent reports back—creating a PR, posting a comment, or sending a Slack message.
96+
4. **Handoff**: You can intervene at any time using [Roomote Control](/roo-code-cloud/roomote-control) to take manual control of the session.
97+
98+
## Credits & Usage
99+
100+
Cloud Agents consume **Cloud Credits**. Usage is measured based on the model used and the duration of the task.
101+
- **Free Tier:** Includes access to free models for standard tasks.
102+
- **Pro Tier:** Unlocks concurrent agent execution and priority queueing.
103+
104+
[Learn more about Cloud Credits](/roo-code-cloud/credits)

0 commit comments

Comments
 (0)