Relentless is an AI agent orchestrator that runs coding agents like Claude Code, Amp, and OpenCode repeatedly until all tasks are complete. It benefits operations teams by automating complex workflows and integrating with existing tools.
git clone https://github.com/ArvorCo/Relentless.gitRelentless is an AI agent orchestrator that runs coding agents like Claude Code, Amp, and OpenCode repeatedly until all tasks are complete. It benefits operations teams by automating complex workflows and integrating with existing tools.
[{"step":1,"action":"Define the task scope in a YAML configuration file or directly in the prompt. Specify the primary agent type (Claude Code, Amp, or OpenCode), maximum retry attempts, and error handling rules.","tip":"Use the Relentless CLI to validate your configuration before execution: `relentless validate --config workflow.yaml`"},{"step":2,"action":"Set up monitoring and notification channels. Configure Slack, email, or webhook integrations to receive status updates for workflow start, completion, or failure.","tip":"For critical workflows, set up multiple notification channels (e.g., Slack + PagerDuty) to ensure visibility across teams."},{"step":3,"action":"Execute the workflow using the Relentless CLI or API. Monitor the initial run to verify agent behavior and adjust retry logic if needed.","tip":"Start with a dry-run mode (`--dry-run` flag) to preview the agent's actions without making changes to your systems."},{"step":4,"action":"Review the generated outputs and logs. Use the audit trail to identify patterns in failures and optimize agent configurations for future runs.","tip":"Tag workflows with metadata (e.g., `project:acme-api`, `priority:high`) to improve traceability in large-scale deployments."},{"step":5,"action":"Integrate Relentless into your CI/CD pipeline or internal tools. Use webhooks to trigger workflows automatically based on events (e.g., Git pushes, Jira ticket updates).","tip":"For Kubernetes-native workflows, deploy Relentless as a sidecar container in your pods to handle agent orchestration alongside your main application."}]
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/ArvorCo/RelentlessCopy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
Use Relentless to automate the following workflow: [TASK_DESCRIPTION]. Configure the agent to retry on failure with [MAX_RETRIES] attempts and log all actions to [LOG_FILE_PATH]. Include error handling for [SPECIFIC_ERROR_TYPES] and notify [TEAM_CHANNEL] upon completion or failure. Use [AGENT_TYPE] as the primary coding agent.
```json
{
"workflow_id": "relentless_20240515_143022",
"status": "completed",
"total_attempts": 3,
"successful_runs": 2,
"failed_runs": 1,
"errors": [
{
"attempt": 1,
"error_type": "timeout",
"message": "Agent failed to fetch repository metadata after 10 minutes",
"timestamp": "2024-05-15T14:32:15Z"
}
],
"outputs": {
"generated_files": [
{"path": "/repos/acme/api/src/main.py", "size": "4.2KB"},
{"path": "/repos/acme/api/tests/test_main.py", "size": "1.8KB"}
],
"logs": [
"[14:30:25] Initializing Relentless orchestrator for workflow 'api_generation'",
"[14:30:27] Agent 'claude-code' started with prompt: 'Generate REST API skeleton for Acme Corp...'",
"[14:31:45] Agent failed: TimeoutError - Repository not cloned within expected time",
"[14:32:01] Retry attempt #2 initiated with agent 'amp'",
"[14:34:12] Agent 'amp' completed successfully",
"[14:35:00] Verifying generated files against requirements..."
]
},
"notifications": {
"slack_channel": "#devops-alerts",
"message": "✅ Relentless workflow 'api_generation' completed in 4m38s. 2 files generated. See logs: /logs/relentless_20240515_143022.log"
}
}
```
The Relentless orchestrator successfully automated the generation of a REST API skeleton for Acme Corp's new microservice. The initial attempt using Claude Code failed after 10 minutes due to a timeout while cloning the repository, but the system automatically triggered a retry with the Amp agent, which completed the task in 2 minutes and 11 seconds. The generated codebase includes a main.py file with 4.2KB of Python code implementing CRUD endpoints for users and products, along with corresponding test files. All actions were logged to `/logs/relentless_20240515_143022.log`, and the team was notified via Slack. The workflow demonstrates Relentless' ability to handle transient failures gracefully while maintaining audit trails for compliance and debugging purposes.Google's multimodal AI model and assistant
AI assistant built for thoughtful, nuanced conversation
IronCalc is a spreadsheet engine and ecosystem
Customer feedback management made simple
Enterprise workflow automation and service management platform
Automate your spreadsheet tasks with AI power
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan