GitHubRepoForker

Fork a GitHub repository from an issue URL and optionally create an issue-specific branch—the first step in AI agent workflows that fix GitHub issues by modifying code.

Key Features

• Parses the repository owner, name, and issue number directly from the GitHub issue URL.
• Creates a new fork or syncs an existing one with the upstream repository.
• Optionally waits for the fork to become available before continuing, with configurable timeout and polling interval.
• Optionally creates a fix-<issue-number> branch in the fork automatically.
• Resolves a GITHUB_TOKEN secret at runtime, keeping credentials out of the pipeline YAML.
• Also available as GitHubRepoForkerTool for use with Haystack agents.

Configuration

1. Store your GitHub personal access token as a secret named GITHUB_TOKEN. For instructions, see Create Secrets.
2. Drag the GitHubRepoForker component onto the canvas from the Component Library.
3. Click on the component to open the configuration panel.
4. Set auto_sync to True to keep existing forks up to date with the upstream.
5. Set create_branch to True to generate a fix-<issue-number> branch automatically.

Connections

GitHubRepoForker receives a single url (the full GitHub issue URL). It outputs repo (the fork path in owner/repo format) and issue_branch (the name of the created branch, if any). Connect these outputs to GitHubFileEditor or GitHubPRCreator to build a complete issue-resolution pipeline.

Source Code

To check this component's source code, open repo_forker.py in the Haystack Core Integrations repository.

Usage Examples

Basic Configuration

  GitHubRepoForker:
    type: haystack_integrations.components.connectors.github.repo_forker.GitHubRepoForker
    init_parameters:
      github_token:
        type: env_var
        env_vars:
        - GITHUB_TOKEN
        strict: true
      raise_on_failure: true
      auto_sync: true
      create_branch: true
      wait_for_completion: false
      max_wait_seconds: 300
      poll_interval: 2

Using the Component in a Pipeline

This example shows a code-fix agent pipeline that forks a repo from a GitHub issue URL, edits a file in the fork, and opens a pull request.

# haystack-pipeline

components:
  repo_forker:
    type: haystack_integrations.components.connectors.github.repo_forker.GitHubRepoForker
    init_parameters:
      github_token:
        type: env_var
        env_vars:
        - GITHUB_TOKEN
        strict: true
      auto_sync: true
      create_branch: true

  file_editor:
    type: haystack_integrations.components.connectors.github.file_editor.GitHubFileEditor
    init_parameters:
      github_token:
        type: env_var
        env_vars:
        - GITHUB_TOKEN
        strict: true

  pr_creator:
    type: haystack_integrations.components.connectors.github.pr_creator.GitHubPRCreator
    init_parameters:
      github_token:
        type: env_var
        env_vars:
        - GITHUB_TOKEN
        strict: true

connections:
- sender: repo_forker.repo
  receiver: file_editor.repo
- sender: repo_forker.issue_branch
  receiver: file_editor.branch

inputs:
  query:
  - repo_forker.url
  - pr_creator.issue_url

outputs:
  result: pr_creator.result

max_runs_per_component: 100

metadata: {}

Parameters

Inputs

Parameter	Type	Description
url	str	Full GitHub issue URL (for example, https://github.com/owner/repo/issues/123).

Outputs

Parameter	Type	Description
repo	str	Fork path in owner/repo format.
issue_branch	Optional[str]	Name of the created branch (for example, fix-123) if create_branch is True, otherwise None.

Init Parameters

These are the parameters you can configure in Pipeline Builder:

Parameter	Type	Default	Description
github_token	Secret	Secret.from_env_var("GITHUB_TOKEN")	GitHub personal access token used for API authentication.
raise_on_failure	bool	True	If True, raises an exception on errors. If False, returns empty strings.
wait_for_completion	bool	False	If True, polls until the fork is fully available before returning.
max_wait_seconds	int	300	Maximum number of seconds to wait when wait_for_completion is True.
poll_interval	int	2	Seconds between status checks when waiting for fork completion.
auto_sync	bool	True	If True, syncs an existing fork with the upstream repository instead of creating a new one.
create_branch	bool	True	If True, creates a fix-<issue-number> branch in the fork.

Related Information

• GitHubFileEditor
• GitHubPRCreator
• GitHubIssueViewer
• Pipeline Components