Avenox Studio — produce brand-locked motion graphics with HyperFrames and composite them onto a cut video. Use when making cutaway/overlay graphics: title cards, diagrams, data-viz, loops, lower-thirds, outro. Proven cutaway + composite recipe with gotchas. Triggers: "make graphics", "add an animation", "motion graphic", "intro/outro card", "data-viz". Part of avenox-video step 3.
Avenox Studio — transcript-driven rough cut (silence + flub/retake removal). Use when cutting a raw talking-head or screen recording: remove dead air AND bad takes/restarts. Proven recipe with real gotchas baked in. Triggers: "rough cut", "cut the silences", "remove flubs/retakes", a raw screen recording to trim. Part of avenox-video step 2.
Generate high-CTR YouTube thumbnails for a mascot-driven channel using parallel gpt-image-2 jobs via the Codex CLI. Use whenever the user asks for a thumbnail, cover, or thumbnail alternatives for a video, wants their mascot on a cover, or asks to retrofit new thumbnails onto old videos. Produces on-model mascot + a big 2-4 word hook + tool logos + high-contrast covers, saved 1280x720 ready for YouTube A/B testing.
Avenox Studio — local-first YouTube video production pipeline (ROUTER, read first). Use for ANY request to edit, cut, produce, assemble, caption, score, or render a long-form video, or to make motion graphics for one. Wraps open-source tooling (auto-editor, mlx-whisper, MLT/melt, ffmpeg) plus HyperFrames for animated graphics. Triggers: "edit this video", "cut the recording", "make graphics", "extract captions", "render the final", a job name, or anything about the video pipeline. The human is director/quality gate; the agent is the operator.
Multi-chain block explorer queries via Etherscan V2 unified API + Foundry cast fallbacks. Look up contract ABI/source, transactions, receipts, event logs, balances, token info, contract creation, gas prices, verify contracts across Ethereum, BSC, Polygon, Arbitrum, Optimism, Base, Avalanche, Sepolia and 60+ other chains. One API key, one endpoint, any chain. Triggers on: etherscan, bscscan, polygonscan, arbiscan, basescan, optimism explorer, snowtrace, scan API, block explorer, contract lookup, get ABI, get source code, verify contract, tx lookup, receipt lookup, event logs, contract creation, contract creator, token holders, gas oracle, abi fetch.
Standalone Codex CLI runner + fleet orchestrator. Does THREE things and always EXECUTES them (never just describes): (1) general code tasks via `codex exec`, (2) high-quality image generation via Codex's built-in `gpt-image-2` tool, and (3) parallel multi-lane fleets — spawning many `codex exec` delegates at once with worktree isolation. Defaults locked: model `gpt-5.6-sol`, reasoning `high`, `--skip-git-repo-check` always. For multiple independent jobs, fire them ALL in parallel — compute is not the constraint, throughput is. Triggers on: "use codex", "run codex", "codex exec", "imagegen", "generate image", "make image", "render this", "ask codex to ...", "have codex ...", "spawn a fleet", "parallel codex", any image-asset request (icons/sigils/banners/portraits/backgrounds/sprites/UI assets/mockups/photoreal/etc.), and any request to delegate code-level work to Codex.
Delegation policy for a multi-model agent stack where the main loop runs on a scarce top-tier model (e.g. Claude Fable) and cheaper/unlimited tiers do the mechanical work. The main loop owns architecture and judgment; sub-agents (Opus) and Codex lanes execute. Load whenever spawning sub-agents (Agent tool, Workflow agent() calls, codex fleets) or planning any delegation.
Run the research/audit handoff loop between an agentic coding harness and a slow, non-agentic frontier reasoning model (GPT Pro on the ChatGPT web UI, or equivalent) with the codebase pre-loaded as split zip bundles. Author a high-signal prompt, then later take the returned report, verify every finding against the LIVE codebase, and implement what holds up. Use whenever the operator wants to send something to a deep external model, prep a prompt or report, scope an issue for deep research, bug-hunting, or optimization analysis, OR when a report comes back and needs checking against the code before anything is implemented. This is the WORKFLOW skill — to build or refresh the zip bundles themselves, use the separate `gptpro` export skill.