Cursor rule Cursor
You are working on Vrooli, an innovative web application that orchestrates AI agents for complex tasks. Follow these guidelines to ensure cohesive execution and continuous process improvement.
Cursor rule Cursor
This document outlines the styling and content guidelines for creating documentation within the Vrooli project. Adhering to these guidelines will ensure consistency, clarity, and maintainability across all project documentation.
Cursor rule Cursor
Vrooli Intelligence System - Core Instructions.
Cursor rule Cursor
Open-source, self-improving autonomous agent swarm🐇.
Cursor rule Cursor
Open-source, self-improving autonomous agent swarm🐇.
Skill Claude CodeCodex
Implement the recommendations from the attached investigation report. Produce a change report documenting what was applied, verified, and deferred.
Skill Claude CodeCodex
Diagnose why the attached agent run(s) failed. Produce a structured report classifying root causes and recommending specific, actionable fixes.
Skill Claude CodeCodex
Design and maintain scenario APIs as stable, intuitive, domain-aligned contracts. The API is the scenario’s entrypoint for core logic and the primary integration surface for other scenarios and the scenario CLI—treat it as a product.
Skill Claude CodeCodex
Session constraints for architectural improvements. These boundaries ensure structural changes enhance the system without destabilizing functionality.
Skill Claude CodeCodex
Prioritize discovering, validating, and safely handling assumptions in this scenario's code, tests, configuration, and interactions.
Skill Claude CodeCodex
Session constraints for audits and investigations. These boundaries ensure thorough analysis without unintended changes.
Skill Claude CodeCodex
Prioritize ensuring that each part of the scenario is only doing the work it is responsible for.
Skill Claude CodeCodex
Manage the full branding lifecycle for Vrooli scenarios — check status, generate assets, apply brands, and export packages using the brand-manager CLI.
Skill Claude CodeCodex
Reference for using Browser Automation Studio (BAS) to execute browser workflows, validate UI behavior, and debug automation failures.
Skill Claude CodeCodex
Session constraints for bug fixes. These boundaries ensure fixes are targeted and don't introduce new issues.
Skill Claude CodeCodex
Integrate scenarios/{{TARGET}}/ into the Vrooli Business Suite bundle so that it participates in subscription gating, credit-based metering, and authenticated downloads managed by the Landing Page Business Suite (LPBS) scenario.
Skill Claude CodeCodex
Create marketing content organized by campaign — generate blog posts, social media copy, email drafts, and ad copy using AI with document-grounded context.
Skill Claude CodeCodex
Audit agent files (SOUL.md, AGENTS.md, TOOLS.md) to identify embedded methodologies that should be extracted into reusable skills. This methodology covers the first half of the promotion pipeline: Agent instructions → Skills. Once capabilities are extracted into skills, they enter the meta-optimization layer where…
Skill Claude CodeCodex
Prioritize making the scenario easier and safer to evolve over time.
Skill Claude CodeCodex needs its repo
Design and maintain scenario CLIs as thin, professional wrappers over the API. The CLI is the developer's primary interface to the scenario—treat it as a product with the same care given to the API itself.
Skill Claude CodeCodex
Prioritize systematic removal of dead code, deprecated implementations, and backwards-compatibility cruft across this scenario. Do not modify code in packages/ (shared code may have external consumers). Do not remove code without verification; all removals must be provably safe.
Skill Claude CodeCodex
Prioritize making the code easier to understand, reason about, and safely modify.
Skill Claude CodeCodex
Prioritize unifying the language of the scenario so that the same concept is referred to consistently across code, UI, tests, and documentation.
Skill Claude CodeCodex
Prioritize designing and refining this scenario’s control surface: the small set of meaningful, safe levers (config, options, flags, parameters) that shape its behavior.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: