Run autonomous SURVEY, DRIVE, INSPECT, STOW, and STANDBY cycles on developer tasks.
npx skills add https://github.com/epologee/leclause-skills --skill rover-epologee
Please help me install this Agent Skill. Skill: rover Source: https://github.com/epologee/leclause-skills/tree/main/packages/autonomous/skills/rover Command: npx skills add https://github.com/epologee/leclause-skills --skill rover-epologee
The rover automates end-to-end task execution for development and documentation work when the operator cannot or should not make every decision, reducing round-trip delays and ensuring deliverables meet predefined Done criteria before handoff.
Invoke the rover with a mission description, a GitHub issue URL, or a loop file path, for example: /autonomous:rover "build the settings page".
High-intent search queries and answers about installing and using this skill.
Autonomous task execution is handled by dispatching an agent to run SURVEY, DRIVE, INSPECT, STOW, and STANDBY cycles on a developer mission. You provide a mission brief via free-form text, a GitHub issue URL, or a loop file, and the agent independently plans, executes, and logs the work.
An autonomous coding agent loop cycles through planning and execution phases to complete development tasks independently. It ensures quality by applying plan-vs-dispatch checks, maintaining a decision audit trail, and requiring a contrarian pride review gate before any artifact leaves the loop.
Recurring autonomous development tasks are scheduled using cron-driven iteration. The agent creates and maintains a loop file in the .autonomous/ directory, schedules a cron job for iterative execution, and supports wake and stop controls for continuous task processing.
Autonomous code generation applies to both code and documentation workflows. The operator provides a mission brief, and the agent autonomously executes the required planning, implementation, and audit logging to meet predefined Done criteria before handoff.
Bad decisions during task execution are mitigated by a decision subsystem that writes an audit trail for every choice. The agent also enforces a contrarian pride review as a hard gate, ensuring deliverables are verified against plan-vs-dispatch checks before finalizing.
No external dependencies are required to run an autonomous rover loop for developer tasks. The agent operates independently, managing its own loop file logging and cron-driven iteration to deliver verified work based on the provided mission brief.
marcusgollAdd formatted development tasks to the deep loop task queue.
semicolon-devteamAutomates acne acceptance criteria supplementation requests for GitHub/Slack issues and updates project statuses via GraphQL/API workflows.
BryceEWatsonCoordinate multi-agent review loops with lint, test, and build checks.
Aiden-KwakCoordinate Claude Code and Codex for planning, implementation, validation, and review.
