Drive local Codex CLI instances non-interactively from an external orchestrator, including unattended launch permissions, working-directory and input handling, configuration and local-model selection, attachments, rollout liveness and token accounting, exit interpretation, and recovery through resume. Use when another…
Generate sound effects and music on ComfyUI — Stable Audio for SFX/ambience and ACE-Step for songs. Verified node graphs, sampler settings, and the missing-text-encoder trap. Use when a project needs generated audio (weapon sounds, footsteps, ambience, stings, music beds); pairs with the comfyui engine skill.
Generate images on ComfyUI with the Krea2 models (krea2turbo, krea2raw) — the exact verified node graph, sampler settings and the two traps that silently ruin output. Use when generating images, game textures, or reference art on ComfyUI and Krea2 is the chosen model; pairs with the comfyui engine skill.
Drive a local ComfyUI instance headlessly from an agent — discover installed models and nodes, build API-format graphs, queue jobs, poll for results, and pull images/audio/video to disk. Includes the discovery protocol for learning a model you have never used (pull its template, never guess the graph) and the failure…
Lets an agent drive headless Grok CLI (grok build) instances toward a user's goal — launch/monitor/salvage non-interactive grok -p runs, configure models and local OpenAI-compatible providers, attach reference files via --prompt-json, and read token usage from JSON output. Use when the user says things like "use grok…
The meta-skill above the per-engine adapters — pick the right coding-agent harness for a goal, drive fleets of headless instances through existing adapters (omp, codex, grok), and when a harness has NO adapter yet, LEARN it empirically and have the harness write its own adapter skill. Use when the user names a harness…
Track which models actually succeed at which kinds of task, and swap one out when the record says it is failing. Records every delegated lane's outcome to a local ledger, surfaces per-model-per-task-type success rates, and recommends a replacement when a model repeatedly fails a category. Use when picking a model for…
Enforce clean human-only authorship on everything git and GitHub — no AI/agent attribution anywhere. Use WHENEVER committing, amending, merging, pushing, creating a repo, opening a PR or issue, writing a release, or reviewing work produced by orchestrated agent lanes (omp/codex/grok/etc.) before it lands in git.
Lets an agent drive local omp (oh-my-pi) instances toward a user's goal — spin up one or many model "lanes", configure providers, decompose the goal into bounded turns, launch/monitor/salvage non-interactive omp runs, verify independently, and account tokens/spend. Use when the user says things like "use omp to have…
Prove a frontend performance change with paired, measured before/after numbers instead of projections. Builds both git refs, serves each production build locally, drives a real signed-in browser N times per condition, and computes a verdict that fails on regression. Use when a change claims a performance benefit, when…
The evidence standard for any QA work — visual, functional, regression, performance, accessibility, or mobile. Establishes what counts as proof, how to capture a baseline BEFORE changing anything, how to pair before/after when state drifts, and how to write a verdict that does not overclaim. Use whenever verifying a…