mmuhlariholdings/Hlavi Agent
Autonomously completes Hlavi tasks marked for agent execution using Claude or any supported AI model
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| model | Model to use. The provider is auto-detected from the model name: Anthropic: claude-opus-4-0, claude-sonnet-4-0, claude-3-5-sonnet-20241022 OpenAI: gpt-4o, gpt-4o-mini, o3, o4-mini Google: gemini-2.0-flash, gemini-1.5-pro | no | claude-opus-4-0 |
| anthropic_api_key | Anthropic API key — required for Claude models | no | — |
| openai_api_key | OpenAI API key — required for GPT / o-series models | no | — |
| google_api_key | Google AI API key — required for Gemini models | no | — |
| hlavi_dir | Path to .hlavi directory, relative to the workspace root | no | .hlavi |
| max_iterations | Maximum agentic turns per task before giving up | no | 50 |
| dry_run | If true, log planned actions without writing files or committing | no | false |
| git_user_name | Git author name used for agent commits | no | Hlavi Agent |
| git_user_email | Git author email used for agent commits | no | agent@hlavi.app |
Outputs
| name | description |
|---|---|
| tasks_processed | Number of autonomous tasks found and attempted |
| tasks_completed | Number of tasks successfully moved to review |