anhhai680/Augment Agent
AI-powered code assistance for GitHub pull requests using Augment's intelligent development tools
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| augment_session_auth | Augment session authentication JSON. Alternative to augment_api_token and augment_api_url. Store as repository secret for security. | no | — |
| augment_api_token | API token for Augment services. Store as repository secret for security. Alternative to augment_session_auth. | no | — |
| augment_api_url | URL endpoint for Augment API requests. Store as repository variable. Alternative to augment_session_auth. | no | — |
| github_token | GitHub token for API access. Must have 'repo' and 'user:email' scopes. | no | — |
| instruction | Direct instruction text for the AI agent. Use this for simple, inline instructions. Mutually exclusive with instruction_file and template inputs. | no | — |
| instruction_file | Path to file containing detailed instructions for the AI agent. Use this for complex, multi-line instructions. Mutually exclusive with instruction and template inputs. | no | — |
| template_directory | Path to directory containing template files. Use this for template-based instruction generation. Mutually exclusive with instruction and instruction_file. | no | — |
| template_name | Name of the template file to use (default: prompt.njk). Only used with template_directory. | no | prompt.njk |
| pull_number | Pull request number for extracting PR context. Required when using templates with PR data. | no | — |
| repo_name | Repository name in format 'owner/repo'. Required when using templates with PR data. | no | — |
| custom_context | Additional context data as JSON string to pass to templates. | no | — |
| model | Model to use; forwarded to auggie as --model | no | — |
| platform | Platform to use for context extraction (github or azure-devops) | no | github |
| azure_devops_token | Azure DevOps Personal Access Token (PAT). Store as repository secret for security. | no | — |
| azure_devops_organization | Azure DevOps organization name | no | — |
| azure_devops_project | Azure DevOps project name | no | — |
| azure_devops_repository | Azure DevOps repository name | no | — |
| azure_devops_pull_request_id | Azure DevOps pull request ID for context extraction | no | — |
| azure_devops_work_item_id | Azure DevOps work item ID for context extraction | no | — |
| azure_devops_build_id | Azure DevOps build ID for context extraction | no | — |
| llm_provider | LLM provider to use (auggie, openai, claude, google) | no | auggie |
| llm_api_key | API key for the selected LLM provider. Store as repository secret for security. | no | — |
| llm_base_url | Base URL for the LLM provider API (for custom endpoints) | no | — |
| llm_temperature | Temperature setting for LLM generation (0.0 to 2.0) | no | 0.7 |
| llm_max_tokens | Maximum tokens for LLM response | no | 4000 |
| llm_timeout | Timeout for LLM API requests in milliseconds | no | 30000 |
| post_comment | Whether to post the response as a PR comment (true/false) | no | false |
| comment_type | Type of comment to post (comment, review) | no | comment |
| review_event | Review event type when comment_type is 'review' (COMMENT, APPROVE, REQUEST_CHANGES) | no | COMMENT |
| use_inline_comments | Whether to post inline comments on specific lines (true/false) | no | false |
| inline_comment_strategy | Strategy for inline comments (review_with_comments, individual_comments) | no | review_with_comments |
Outputs
no outputs