Skip to content

Conversation

@jeffnash
Copy link
Contributor

Summary

Adds comprehensive test coverage for the Copilot integration.

Changes

  • copilot_headers_test.go: Tests for agent detection, X-Initiator header logic, initiator persistence, and vision headers
  • copilot_executor_test.go: Tests for model prefix stripping, token fetching, account type precedence, and refresh behavior

Stack

This is PR 5/5 in the Copilot feature stack:

  1. Infrastructure & Config (feat(copilot): add shared infrastructure and config [1/5] #380)
  2. Auth Flow (feat(copilot): implement GitHub Copilot authentication flow [2/5] #381)
  3. Executor & Models (feat(copilot): add Copilot request executor and model registry [3/5] #382)
  4. Gemini Reasoning (feat(copilot): add Gemini 3 Pro reasoning support [4/5] #383)
  5. [This PR] Tests

Depends on: #383 - merge that first

- Add auth directory management helper (GetAuthDir)
- Add random hex string generator for request IDs
- Add helper for generating unique machine identifiers
@chatgpt-codex-connector
Copy link

Codex usage limits have been reached for code reviews. Please check with the admins of this repo to increase the limits by adding credits.
Credits must be used to enable repository wide code reviews.

@gemini-code-assist
Copy link

Summary of Changes

Hello @jeffnash, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request represents the final phase (5/5) of the GitHub Copilot feature stack, focusing on solidifying the integration with comprehensive test coverage. It introduces the full implementation of the Copilot authentication flow, API request handling, and model management, alongside robust unit tests to ensure the reliability and correctness of the new functionality.

Highlights

  • Copilot Integration Core: Introduced core components for GitHub Copilot integration, including dedicated authentication services, API configuration, and error handling.
  • Authentication Flow: Implemented the OAuth2 device code flow for Copilot authentication, supporting both CLI (-copilot-login) and management API (/copilot-auth-url) initiation.
  • Executor and Model Management: Added a CopilotExecutor to handle API requests, manage token refresh, strip model prefixes, and dynamically fetch available Copilot models.
  • Dynamic Header Generation: Implemented logic for dynamically setting Copilot-specific HTTP headers, such as X-Initiator (for agent detection and persistence) and Copilot-Vision-Request.
  • Gemini Reasoning Cache: Introduced a caching mechanism for Gemini reasoning fields to ensure proper tool call functionality in Copilot requests.
  • Comprehensive Test Coverage: Added extensive unit tests for the Copilot executor, header generation logic, token management, and account type validation.
Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature Command Description
Code Review /gemini review Performs a code review for the current pull request in its current state.
Pull Request Summary /gemini summary Provides a summary of the current pull request in its current state.
Comment @gemini-code-assist Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help /gemini help Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

  1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution.

Copy link

@gemini-code-assist gemini-code-assist bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request introduces a comprehensive integration for GitHub Copilot, including the complex device code authentication flow, management API endpoints, and the request executor. The implementation is very thorough, with good error handling, caching mechanisms, and support for various Copilot features like different account types and vision models. The addition of extensive tests for the new functionality is also a great contribution.

I've found a couple of critical copy-paste errors that would prevent compilation, and a minor issue in the example configuration. After addressing these, the feature should be in great shape.

Comment on lines 219 to 247
// CopilotKey represents the configuration for GitHub Copilot API access.
// Authentication is handled via device code OAuth flow, not API keys.
type CopilotKey struct {
// AccountType is the Copilot subscription type (individual, business, enterprise).
// Defaults to "individual" if not specified.
AccountType string `yaml:"account-type" json:"account-type"`

// ProxyURL overrides the global proxy setting for Copilot requests if provided.
ProxyURL string `yaml:"proxy-url,omitempty" json:"proxy-url,omitempty"`

// AgentInitiatorPersist, when true, forces subsequent Copilot requests sharing the
// same prompt_cache_key to send X-Initiator=agent after the first call. Default false.
AgentInitiatorPersist bool `yaml:"agent-initiator-persist" json:"agent-initiator-persist"`
}

// CopilotKey represents the configuration for GitHub Copilot API access.
// Authentication is handled via device code OAuth flow, not API keys.
type CopilotKey struct {
// AccountType is the Copilot subscription type (individual, business, enterprise).
// Defaults to "individual" if not specified.
AccountType string `yaml:"account-type" json:"account-type"`

// ProxyURL overrides the global proxy setting for Copilot requests if provided.
ProxyURL string `yaml:"proxy-url,omitempty" json:"proxy-url,omitempty"`

// AgentInitiatorPersist, when true, forces subsequent Copilot requests sharing the
// same prompt_cache_key to send X-Initiator=agent after the first call. Default false.
AgentInitiatorPersist bool `yaml:"agent-initiator-persist" json:"agent-initiator-persist"`
}

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

critical

The CopilotKey struct is declared multiple times, which will cause a compilation error. Please remove the duplicate declarations.

}
accountType = parsed
}
}

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

critical

This closing brace is extra and will cause a compilation error. Please remove it.

Comment on lines 110 to 149
# GitHub Copilot account configuration
# Note: Copilot uses OAuth device code authentication, NOT API keys or tokens.
# Do NOT paste your GitHub access token or Copilot bearer token here.
# Tokens are stored only in auth-dir JSON files, never in config.yaml.
#
# To authenticate:
# - CLI: run with -copilot-login flag
# - Web: use the /copilot-auth-url management endpoint
#
# After OAuth login, tokens are managed automatically and stored in auth-dir.
# The entries below only configure account type and optional proxy settings.
#copilot-api-key:
# - account-type: "individual" # Options: individual, business, enterprise
# proxy-url: "socks5://proxy.example.com:1080" # optional: proxy for Copilot requests

# # When set to true, this flag forces subsequent requests in a session (sharing the same prompt_cache_key)
# # to send the header "X-Initiator: agent" instead of "vscode". This mirrors VS Code's behavior for
# # long-running agent interactions and helps prevent hitting standard rate limits.
# agent-initiator-persist: true

# GitHub Copilot account configuration
# Note: Copilot uses OAuth device code authentication, NOT API keys or tokens.
# Do NOT paste your GitHub access token or Copilot bearer token here.
# Tokens are stored only in auth-dir JSON files, never in config.yaml.
#
# To authenticate:
# - CLI: run with -copilot-login flag
# - Web: use the /copilot-auth-url management endpoint
#
# After OAuth login, tokens are managed automatically and stored in auth-dir.
# The entries below only configure account type and optional proxy settings.
#copilot-api-key:
# - account-type: "individual" # Options: individual, business, enterprise
# proxy-url: "socks5://proxy.example.com:1080" # optional: proxy for Copilot requests

# # When set to true, this flag forces subsequent requests in a session (sharing the same prompt_cache_key)
# # to send the header "X-Initiator: agent" instead of "vscode". This mirrors VS Code's behavior for
# # long-running agent interactions and helps prevent hitting standard rate limits.
# agent-initiator-persist: true

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

These configuration blocks for copilot-api-key are duplicates of the one above. Please remove them to avoid confusion.

"client_id": GitHubClientID,
"scope": GitHubAppScopes,
}
bodyBytes, _ := json.Marshal(reqBody)

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The error returned by json.Marshal is being ignored. While it's unlikely to fail with a map[string]string as input, it's good practice to handle potential errors. This also applies to a similar case on line 160.

bodyBytes, err := json.Marshal(reqBody)
if err != nil {
	return nil, fmt.Errorf("%w: failed to marshal request body: %v", ErrDeviceCodeFailed, err)
}

- Add Copilot section with agent-initiator-persist flag
- Add scanner buffer size configuration
- Add account type configuration option
- Add copilot types with account type validation
- Document configuration options in example config
- Add tests for util and copilot types
- Add device-code OAuth flow with GitHub token exchange
- Implement Copilot token acquisition and refresh logic
- Add account type handling (individual/business/enterprise)
- Add token persistence and storage management
- Add CLI login command (cliproxy-api copilot login)
- Register Copilot refresh handler in SDK
- Validate account_type with warning for invalid values
- Implement Copilot executor with header injection
- Add VS Code version headers and integration IDs
- Add agent header logic (X-Initiator detection)
- Add vision request header for image inputs
- Add static model registry with raptor-mini and oswe-vscode-prime
- Add management API endpoints for auth files

Note: Full model list and dynamic fetching added in gemini branch
…tching

- Add full Copilot model registry (all GPT, Claude, Gemini, Grok models)
- Implement dynamic model fetching from Copilot API with caching
- Add Gemini reasoning capture and injection for tool calls
- Add reasoning_opaque and reasoning_text handling for Gemini 3 models
- Evict model and reasoning caches on auth removal
- Add 30-second retry delay for 429 quota errors

Credit: Reverse engineering insights adapted from github.com/aadishv/vscre
- Add executor tests for token handling and refresh logic
- Add header tests for X-Initiator and vision detection
- Add tests for account type validation and HTTP status errors
- Add tests for model registry and alias generation
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants