Manage AI research run execution policies, interaction modes, and safety allowances.
npx skills add https://github.com/TenureAI/PhD-Zero --skill run-governor
Please help me install this Agent Skill. Skill: run-governor Source: https://github.com/TenureAI/PhD-Zero/tree/main/.agents/skills/run-governor Command: npx skills add https://github.com/TenureAI/PhD-Zero --skill run-governor
This Skill ensures that AI research runs are consistent, auditable, and mode-aware by managing execution policies, interaction modes, and safety protocols.
full-auto, moderate, or detailed interaction modes.Use the run governor to initialize a new research run, setting the interaction mode to 'moderate' and confirming the execution target.
High-intent search queries and answers about installing and using this skill.
To manage research run execution policies, you can define interaction modes, enforce safety allowances, and handle high-resource action prompts. This ensures your AI research tasks remain consistent, auditable, and mode-aware throughout the run lifecycle.
Available interaction modes for AI research workflow execution include full-auto, moderate, and detailed. These modes define user interaction frequency, governing how the AI agent controls the research run and decides between continuing or starting new runs.
Setting up a new AI research run involves creating unique run identifiers and defining directory structures for logging and output. You initialize the run by selecting an interaction mode and confirming the execution target.
Yes, the run governor supports execution target selection by determining whether research runs execute locally or remotely. It manages the necessary configurations for the chosen execution target during run initiation.
Safety protocols should be used to handle high-resource actions whenever an AI agent control task requires strict policy enforcement. The run governor enforces safety allowances and prompts user confirmation for high-resource actions during research execution.
louzhengshuaiFrame research objectives, datasets, metrics, and baselines.
growthxaiValidate and fix Output SDK workflow folder structures.
GrinRusExecute and audit full AIDD workflow runs against pinned public GitHub repositories.
nocooRun predefined evaluation cases against prompts to validate agent behavior and generate Markdown reports.