realemmetts

by Emmett Shaughnessy

A local task board and project memory for Codex. Plan work, track dependencies and completion checks, and resume in a new chat with the context intact. Seven portable skills also support Claude Code and capability-aware ChatGPT handoffs.

by Emmett Shaughnessy

A local task board and project memory for Codex. Plan work, track dependencies and completion checks, and resume in a new chat with the context intact. Seven portable skills also support Claude Code and capability-aware ChatGPT handoffs.

by Emmett Shaughnessy

Emmett Shaughnessy's personal Codex skills bundle. Provides SHAUGHV skills for self-routing long-horizon agentic prompt engineering and cross-runtime subagent brief compilation across software, data, mathematics, and science; falsifiable contracts, evidence receipts, Claude Fable 5 / Opus 5 and GPT-5.6 Sol / Codex overlays, prompt evaluation, loop escape, convergence recovery, evidence-aware retries, progressive slicing, critical thinking, OpenAI audio, PreText, Quiver AI, Mistral AI, brand design, animated brandmark, human changelog, CDN, naming, Google Cloud Storage, bug triage, design patterns, evidence-backed debugging, defensive programming, git workflow, security audit/red-team, typed session handoffs, iterative milestone planning, proportional workflow optimization, image generation/upscaling, deliberate learning, logical reasoning, personal productivity, optional MCP connection guidance, wb300, usage status line, and TT;DR. (The Claude-only subagent model/effort preference skill is omitted from Codex. The task + workplace-memory system lives in shaughv-tasks.)

by Emmett Shaughnessy

Emmett Shaughnessy's personal Codex skills bundle. Provides SHAUGHV skills for self-routing long-horizon agentic prompt engineering and cross-runtime subagent brief compilation across software, data, mathematics, and science; falsifiable contracts, evidence receipts, Claude Fable 5 / Opus 5 and GPT-5.6 Sol / Codex overlays, prompt evaluation, loop escape, convergence recovery, evidence-aware retries, progressive slicing, critical thinking, OpenAI audio, PreText, Quiver AI, Mistral AI, brand design, animated brandmark, human changelog, CDN, naming, Google Cloud Storage, bug triage, design patterns, evidence-backed debugging, defensive programming, git workflow, security audit/red-team, typed session handoffs, iterative milestone planning, proportional workflow optimization, image generation/upscaling, deliberate learning, logical reasoning, personal productivity, optional MCP connection guidance, wb300, usage status line, and TT;DR. (The Claude-only subagent model/effort preference skill is omitted from Codex. The task + workplace-memory system lives in shaughv-tasks.)