{"id":51081236,"url":"https://github.com/tumblecat44/universe-grok","last_synced_at":"2026-06-23T18:03:18.971Z","repository":{"id":359270411,"uuid":"1242351948","full_name":"tumblecat44/universe-grok","owner":"tumblecat44","description":"Truth-seeking orchestration harness for Grok Build. Actively researches current best practices at session start via X and web.","archived":false,"fork":false,"pushed_at":"2026-05-21T05:05:57.000Z","size":105,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-05-21T11:53:51.516Z","etag":null,"topics":["agent","ai-agents","grok","grok-build","harness","orchestration","subagents","truth-seeking","xai"],"latest_commit_sha":null,"homepage":null,"language":"Shell","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/tumblecat44.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":"GOVERNANCE.md","roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":"AGENTS.md","dco":null,"cla":null}},"created_at":"2026-05-18T10:53:46.000Z","updated_at":"2026-05-21T05:06:00.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/tumblecat44/universe-grok","commit_stats":null,"previous_names":["tumblecat44/universe-grok"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/tumblecat44/universe-grok","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tumblecat44%2Funiverse-grok","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tumblecat44%2Funiverse-grok/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tumblecat44%2Funiverse-grok/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tumblecat44%2Funiverse-grok/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/tumblecat44","download_url":"https://codeload.github.com/tumblecat44/universe-grok/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tumblecat44%2Funiverse-grok/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34700915,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-23T02:00:07.161Z","response_time":65,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["agent","ai-agents","grok","grok-build","harness","orchestration","subagents","truth-seeking","xai"],"created_at":"2026-06-23T18:03:18.059Z","updated_at":"2026-06-23T18:03:18.964Z","avatar_url":"https://github.com/tumblecat44.png","language":"Shell","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Universe Grok\n\n**A truth-seeking orchestration harness for Grok Build.**\n\nUniverse Grok turns Grok Build from a powerful model into a **system that actively pursues current truth** instead of relying on stale training data.\n\nIt is not just another skill. It is an attempt to build the highest-leverage way to use Grok for serious software engineering work.\n\n## Core Philosophy\n\n\u003e The biggest limitation of current coding agents is not intelligence — it is operating on outdated internal knowledge.\n\nUniverse Grok's defining behavior:\n\n- At the start of every session, it proactively researches **current best practices** using X (Twitter) and Web search.\n- It treats \"what the model already knows\" as a hypothesis, not as truth.\n- All orchestration, verification, and agent collaboration exists to serve **better contact with reality**.\n\nSee [CONSTITUTION.md](./CONSTITUTION.md) for the full principles and Level 2 → Level 3 vision.\n\n## Current Status\n\n- **Target**: Level 2 (Reliable, high-quality harness)\n- **Vision**: Level 3 (The standard way serious people use Grok Build)\n\nWe are building this in the open as a public project.\n\n## Features (Current)\n\n- Strong curated personas (`universe-verifier`, `universe-researcher`, `universe-tactician`, etc.)\n- Session Start automatic best-practice research hook (X + Web)\n- **First Questions** companion skill (`/first-questions`) — ruthless \"Question every requirement\" ritual that forces a one-sentence goal and deletes scope before any work begins\n- **First Principles** companion skill (`/first-principles`) — forces explicit sourced bedrock facts and reconstruction from reality (not analogy) before solution work; co-equal highest-authority artifact with first-questions.md\n- Clear Constitution that actually influences behavior\n- Deep integration philosophy with existing high-quality skills (`/check`, `/best-of-n`, etc.)\n\n## Installation\n\n```bash\n# 1. Clone the repo\ngit clone https://github.com/your-org/universe-grok.git ~/.grok/plugins/universe-grok\n\n# 2. Install the personas (recommended)\ncp ~/.grok/plugins/universe-grok/personas/*.toml ~/.grok/personas/\n\n# 3. Restart Grok Build or reload plugins\n```\n\nAfter installation, activate with natural language:\n- \"universe mode\" / \"full power\"\n- \"first questions\" + \"first principles\" (the required non-negotiable pair as the very first steps on any serious work)\n- \"first principles\" (the required reconstruction step after first-questions, before orchestration)\n- \"push Grok to the limit\"\n\nRecommended order for maximum-leverage sessions:\n1. `/first-questions` — Kill ambiguity and delete everything that fails the test.\n2. `/first-principles` (on surviving scope) — Reconstruct the minimal solution from explicit bedrock truths before any design or implementation.\n3. research hook if needed\n4. `/universe` — Orchestrate with the two sacred artifacts (first-questions.md + first-principles.md) as co-equal contracts.\n\nOr explicitly with `/universe` and `/first-questions`.\n\n## How It Works\n\n**The highest-leverage flow** (use this):\n\n1. **First Questions** (`/first-questions`) — Before anything else, force the One-Sentence Goal and ruthlessly interrogate every requirement using Elon’s \"Question every requirement\" protocol. Delete what fails. Produce the sacred `.grok/first-questions/first-questions.md` artifact.\n2. **Session Start Research** — Universe Grok’s hook triggers fresh X + web research into current best practices.\n3. **Orchestration** — Uses strong personas and optimal subagent topologies.\n4. **Verification** — Independent `universe-verifier` on every meaningful change.\n\nFirst Questions is the foundation. Universe Grok orchestration is only as good as the clarity of the goal it receives.\n\n## Project Structure\n\n```\nuniverse-grok/\n├── skills/\n│   ├── universe-grok/SKILL.md        # Main orchestration logic\n│   └── first-questions/SKILL.md      # Ruthless requirement interrogation (run this first)\n├── personas/                         # Universe-optimized personas\n├── hooks/                            # SessionStart truth-seeking hook\n├── CONSTITUTION.md                   # The soul of the project\n├── plugin.toml\n└── README.md\n```\n\n## Roadmap\n\nSee [CONSTITUTION.md](./CONSTITUTION.md) for Level definitions.\n\nHigh priority:\n- Stronger Session Start research protocol\n- Better automatic stack detection\n- More powerful personas\n- Richer research artifact consumption by subagents\n\n## Contributing\n\nWe welcome contributions, especially in:\n- High-signal personas\n- Research protocols and truth-seeking mechanisms\n- Real usage examples and case studies\n\nPlease read [CONTRIBUTING.md](./CONTRIBUTING.md) and [CONSTITUTION.md](./CONSTITUTION.md) first.\n\n## License\n\nMIT\n\n---\n\n**\"The difference between a strong model and a strong system is the relentless pursuit of truth.\"**\n\nThis project is in active development toward Level 3. Star the repo and follow along if you want to push Grok Build to its absolute limits.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftumblecat44%2Funiverse-grok","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ftumblecat44%2Funiverse-grok","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftumblecat44%2Funiverse-grok/lists"}