Generate, audit, and recover nexus-autoloop execution loops with evidence-backed assessments.
npx skills add https://github.com/simota/agent-skills --skill orbit
Please help me install this Agent Skill. Skill: Orbit Source: https://github.com/simota/agent-skills/tree/main/orbit Command: npx skills add https://github.com/simota/agent-skills --skill orbit
This Skill automates the creation, execution, auditing, and recovery of autonomous loops, ensuring tasks run reliably to completion with auditable evidence.
nexus-autoloop from a given goal.Use the Orbit skill to generate a new nexus-autoloop script set for the goal 'Implement user authentication'.
High-intent search queries and answers about installing and using this skill.
To generate autonomous loop scripts for the nexus-autoloop framework, provide a specific goal and the system creates ready-to-run script sets. These scripts include measurable operation contracts with acceptance criteria and resumable state management.
Autonomous loop auditing classifies live loop status and verifies evidence by assessing state integrity. State recovery works by identifying state drift or failures during execution and automatically attempting evidence-backed recovery to ensure reliable task completion.
Yes, you can automate complex multi-step processes that might fail. The system generates execution scripts, defines measurable success conditions, runs the process, and automatically attempts evidence-backed recovery if it encounters failures or state drift.
To design operation contracts with acceptance criteria for automated tasks, define your goal and let the system build measurable loop contracts. These contracts establish resumable state integrity and specific success conditions for reliable long-running automation.
Autonomous loop recovery for long-running tasks relies on evidence-backed assessments and state integrity verification. While it automatically attempts recovery from state drift and failures, complex automated processes require properly defined operation contracts to ensure reliable recovery outcomes.
yu13130122297Identify uncertainties in research framing, datasets, metrics, and baselines.
jchildreeVerify completion claims by executing and confirming command outputs before commits.
z1439527767Detect and recover failed AI agent tool executions with retries and checkpoints.
TeiNamEngineer autonomous loops with goal completion contracts and verification gates in Kiro.