presentation
49Skill Claude CodeCodex
Presentation expert for slide structure, storytelling, visual design, and audience engagement.
Skill Claude CodeCodex
Presentation expert for slide structure, storytelling, visual design, and audience engagement.
Skill Claude CodeCodex
Project management expert for Agile, estimation, risk management, and stakeholder communication.
Skill Claude CodeCodex
Prometheus monitoring expert for PromQL, alerting rules, Grafana dashboards, and observability.
Skill Claude CodeCodex
Prompt engineering expert for chain-of-thought, few-shot learning, evaluation, and LLM optimization.
Skill Claude CodeCodex
Python expert for stdlib, packaging, type hints, async/await, and performance optimization.
Skill Claude CodeCodex
React expert for hooks, state management, Server Components, and performance optimization.
Skill Claude CodeCodex
Redis expert for data structures, caching patterns, Lua scripting, and cluster operations.
Skill Claude CodeCodex
Regular expression expert for crafting, debugging, and explaining patterns.
Skill Claude CodeCodex
Rust programming expert for ownership, lifetimes, async/await, traits, and unsafe code.
Skill Claude CodeCodex
Privacy-respecting metasearch specialist using SearXNG instances.
Skill Claude CodeCodex
Security audit expert for OWASP Top 10, CVE analysis, code review, and penetration testing methodology.
Skill Claude CodeCodex
Sentry error tracking and debugging specialist.
Agent Claude Code
Use to autonomously optimize a model's AMK megakernel, runs the correctness-gated propose->eval->keep/revert loop (and the unattended autoresearch driver) and reports the measured, drift-robust speedup over AMK's own baseline.
Command Claude Code
Launch the unattended / overnight AMK autoresearch driver on $ARGUMENTS (model [gpu] [minutes|iters]).
Command Claude Code
One-shot compile + verify a model into a CUDA megakernel via amk compile on $ARGUMENTS (model [gpu]).
Command Claude Code
Drive an interactive AMK propose -> eval -> keep/revert megakernel schedule session on $ARGUMENTS (model [gpu]).
Skill Claude CodeCodex
Use when optimizing or generating a CUDA megakernel for a HuggingFace Llama-family model with AutoMegaKernel (AMK), drives the correctness-gated propose -> eval -> keep/revert loop (or hands off to the unattended autoresearch driver).
MCP server Claude CodeCodexCursor +2
MCP server "automegakernel" as configured in RightNow-AI/AutoMegaKernel. Runs locally from the agent Python package.
Instructions file CodexOpenCode
Instructions for RightNow-AI/AutoMegaKernel, covering what amk is, the edit surface, scheduleconfig + kernelknobs only, canonical tool surface (mcp), use these exact names, codex mcp server config (/.codex/config.toml) and the copy-paste loop a codex agent runs.
Instructions file
Instructions for RightNow-AI/auto, covering claude.md, auto (the agi compiler), what this is, the three objects, pipeline and stack.
Instructions file CodexOpenCode
Instructions for RightNow-AI/autoevolve, covering autoevolve agent instructions, constitution and scope, what autoevolve does, connect and worker loop.
Instructions file
Instructions for RightNow-AI/autoevolve, covering claude.md — autoevolve, 1. what autoevolve is, the five invariants (the product — never trade these away), 2. product surfaces and 3. repo layout.
Skill Claude CodeCodex
Use this skill whenever asked to join an autoevolve run, evolve code toward a measured target, work an evolution population, mutate a candidate under EVOLVE-BLOCK rules, or report measured autoevolve progress and artifacts.
MCP server Claude CodeCodexCursor +2
Multi-agent research competition orchestrator for autoresearch. Runs locally from the autoevolve Python package.