← all publishers

unsafe9

@unsafe9 source repo

Report this profile Claim this profile

• 6 skills
• 0 followers
• 3 hours ago last updated

1. ▌
   Viz · unsafe9 bundle

   Render a long structured answer — or any data worth visualizing — as a self-contained single-file scroll web page, by writing only thin YAML data (an ordered list of typed blocks) that a build script turns into DOM, big numbers, charts, SVG diagrams, trees, diffs, and layout. You never hand-write HTML or compute coordinates. Strongly prefer this whenever your reply would otherwise be a long expository read the user goes through start-to-finish (explaining how something works, an analysis or investigation writeup, comparing options or tradeoffs, a research/report summary, a codebase or architecture exploration, a multi-step plan, a status/metrics snapshot, a post-mortem) OR whenever the answer carries data with shape — metrics/big numbers, comparison tables, rankings, flow/architecture/sequence diagrams, file trees, code diffs, timelines — or a decision with options to put in front of the user and collect their pick. Also trigger on explicit asks — "visualize this", "diagram this", "show this as a page". Skip
   0 installs
2. ▌
   Rescue · unsafe9 bundle

   Rescue
   0 installs
3. ▌
   Handoff · unsafe9 bundle

   Transfer a self-contained task to a separate, persistent agent surface that keeps working on its own — Codex app (via a `codex://` deeplink), Claude Code in tmux, or Codex CLI in tmux. Composes a self-contained brief, launches the surface, and reports where it is running; control moves to that surface and the work continues independently in its own app or terminal window. The point is to set up the work to live on elsewhere, not to wait on it here. Use when the user wants to continue work in another app or terminal, open or launch a task in Codex, spin up an autonomous background or tmux agent, or says 'handoff', 'continue this in tmux', 'open it in the Codex app', 'run it in the background'. When the user says codex, default to Codex app; use Codex CLI only when explicitly named.
   0 installs
4. ▌
   Instruction · unsafe9 bundle

   Write, edit, or review durable agent instructions in CLAUDE.md, AGENTS.md, skills, agent prompts, or context-injecting hooks. Also use to analyze agent mistakes and propose prevention.
   0 installs
5. ▌
   Generate Image · unsafe9 bundle

   Generate images from a text prompt using OpenAI's gpt-image-2 (default) or Google Gemini (Nano Banana) and open them in the system image viewer. Use whenever the user asks to create, generate, draw, make, or produce an image/picture/illustration/poster/icon/logo/infographic/thumbnail/wallpaper from a description — even if they don't mention OpenAI, Gemini, gpt-image-2, or DALL·E explicitly. Triggers include phrases like 'generate an image of', 'draw me a', 'create a picture', 'make a visual of'. Also use when the user names a backend ('use gemini', 'use gpt') or wants to compare/race both models on the same prompt ('race gpt vs gemini'), and for iterating on a prompt (regenerating with variations or a refined description). Requires OPENAI_API_KEY (and GEMINI_API_KEY for the gemini provider or --race).
   0 installs
6. ▌
   Alfred Workflow · unsafe9 bundle

   Author Alfred 5 workflows as code by writing the native info.plist XML — object graph, connections, canvas layout (uidata), user configuration — then validating, packaging into .alfredworkflow, and installing. Use whenever the user wants to create, modify, lay out, debug, package, or install an Alfred workflow or macOS launcher automation, mentions .alfredworkflow, Script Filter, Alfred keyword/hotkey/external trigger, or says 'build an Alfred workflow', 'add an Alfred keyword'. Trigger even when the user only describes the desired launcher behavior without naming Alfred internals.
   0 installs