Skills from clawhub.ai/okaris

This skill automates AI podcast production using the `infsh` CLI and remote TTS/music apps like `infsh/kokoro-tts` and `openrouter/claude-sonnet-45`. It instructs executing `curl -fsSL https://cli.inference.sh | sh` and multiple `infsh app run` commands, which download/execute remote code and perform network calls.

Dangerous skill: instructs running a remote piped shell `curl -fsSL https://cli.inference.sh | sh` and invoking `infsh` to run external model apps. It is a setup/usage guide for `infsh`-based image generation and links to `https://inference.sh` documentation.

High-risk skill installs and runs a CLI via `curl -fsSL https://cli.inference.sh | sh` and executes `infsh` commands that make network calls and write local files. It documents multi-step AI content pipelines and saves outputs to files like `script.json`.

Dangerous skill: runs a remote installer with `curl -fsSL https://cli.inference.sh | sh` and executes `infsh app run` commands that call remote model services. The skill provides prompts and examples to generate AI product photography.

This skill automates browser interactions via `infsh`/Playwright (functions: `open`, `interact`, `execute`, `screenshot`, `close`). It instructs running `curl -fsSL https://cli.inference.sh | sh`, performs external `infsh` network calls (e.g., `https://inference.sh`) and uses env vars like `APP_USERNAME`/`APP_PASSWORD`, enabling remote execution and credential use.

High-risk skill: provides automation patterns and runnable examples for AI workflows and scheduled jobs. It instructs execution of remote installers (`curl -fsSL https://cli.inference.sh | sh`), running `infsh` commands (including `infsh login`), and sending alerts to `https://your-webhook.com/alert`.

High-risk skill: installs and runs a remote CLI via `curl -fsSL https://cli.inference.sh | sh`. It instructs using `infsh login` and `infsh app run`, which perform network calls to `https://inference.sh` and save inputs to `input.json`.

This skill automates creation of AI marketing videos using the `inference.sh` CLI and `infsh app run` commands. It instructs downloading and executing a remote installer via `curl -fsSL https://cli.inference.sh | sh`, running `infsh login`, and making external network calls to `https://inference.sh` and `https://trending-music.mp3`.

Dangerous skill: downloads and executes a remote installer via `curl -fsSL https://cli.inference.sh | sh` and instructs `infsh login`. It then runs `infsh app run` commands that fetch external resources such as `https://portrait.jpg` and `https://speech.mp3`.

Provides a React/Next.js agent UI component with runtime, tools, streaming, and client-side tools. Includes instructions to run `npx shadcn@latest add https://ui.inference.sh/r/agent.json`, install `@inferencesh/sdk`, and set `INFERENCE_API_KEY` in `.env.local`, which causes CLI installs and storing an API key.

Dangerous skill: installs and runs a remote script via `curl -fsSL https://cli.inference.sh | sh` and invokes networked `infsh` commands; it also instructs setting `INFSH_API_KEY`. This enables remote code execution and external network access.

Dangerous skill: installs and executes remote code via `curl -fsSL https://cli.inference.sh | sh` and runs `infsh app run ...` to invoke external services. It also directs `infsh login`, requiring credentials and making network calls to `https://cli.inference.sh` and various `infsh app run` endpoints.

High-risk skill that automates social media asset generation while instructing installation and execution of a remote CLI. It directs `curl -fsSL https://cli.inference.sh | sh` and `infsh login`, and runs `infsh` commands that enable network calls and credential use.

High-risk skill: runs shell commands that download-and-execute code (`curl -fsSL https://cli.inference.sh | sh`), instructs `infsh login`, and invokes `infsh app run` against external URLs (e.g., `https://your-photo.jpg`). The skill is a background-removal helper using the `infsh` CLI.

This skill documents TTS and voice-cloning workflows using the `infsh` CLI and apps such as `infsh/kokoro-tts`. It instructs downloading and executing a remote installer `curl -fsSL https://cli.inference.sh | sh` and running `infsh login`, which triggers network calls to `https://cli.inference.sh`.

High-risk skill: instructs downloading and executing a remote installer (`curl -fsSL https://cli.inference.sh | sh`) and running the `infsh` CLI. Provides app-store screenshot guidance and many `infsh app run` examples that invoke remote models.

High-risk skill: provides competitor analysis templates and automates research with CLI tooling. It instructs shell execution (`curl -fsSL https://cli.inference.sh | sh`) and remote calls via `infsh app run` (e.g., `https://competitor.com`), enabling code install and external requests.

High-risk skill: a data-visualization guide that instructs installing and running a remote CLI (`curl -fsSL https://cli.inference.sh | sh`) and invoking `infsh` to run remote executors. It also directs an `infsh login` step and network calls to `https://cli.inference.sh`.

This skill creates research-backed customer personas and provides CLI-based examples using `inference.sh`. It instructs running `curl -fsSL https://cli.inference.sh | sh` and `infsh` commands (including `infsh login`), which execute downloaded scripts and perform external network calls.

This skill runs the `infsh` CLI to generate images and includes a remote installer command `curl -fsSL https://cli.inference.sh | sh`. It instructs network calls to `https://cli.inference.sh` and `https://inference.sh/...` and executes `infsh` commands, creating high risk from remote script execution.

This skill documents the `@inferencesh/sdk` JavaScript/TypeScript SDK for running AI apps, agents, file uploads, proxies, and webhooks. It includes explicit instructions to call external URLs like `https://hooks.slack.com/...` and `https://api.github.com/...`, use env vars such as `process.env.INFERENCE_API_KEY`, and execute code via `eval`/`.codeExecution(true)`.

This skill provides LinkedIn post-writing guidance and example CLI workflows for `inference.sh` and `infsh`. It instructs running `curl -fsSL https://cli.inference.sh | sh` and `infsh app run` commands that download/execute remote code and perform network calls.

Dangerous skill: runs shell commands that download-and-execute remote scripts (`curl -fsSL https://cli.inference.sh | sh`) and invokes remote inference services (`https://inference.sh`, `infsh app run`). It provides logo design guidance and `infsh` CLI examples such as `infsh login`.

High-risk skill: installs and executes a remote CLI via `curl -fsSL https://cli.inference.sh | sh` and runs `infsh` commands. It also instructs `infsh login` (credential use) and invokes networked services such as `developers.facebook.com/tools/debug/`, `cards-dev.twitter.com/validator`, and `opengraph.xyz`.

This skill provides guidance for writing press releases in AP style and includes research/fact-checking commands using the `inference.sh` CLI. It instructs running `curl -fsSL https://cli.inference.sh | sh`, executing `infsh` apps (e.g., `infsh app run tavily/search-assistant`), and running `infsh login`.

This skill provides a Python code executor that runs user code via the `infsh` CLI and returns files placed in `outputs/`. It instructs running `curl -fsSL https://cli.inference.sh | sh` and `infsh login`, enabling remote code download/installation and external network access.

This skill documents generating AI videos using the `inference.sh` CLI and provides `infsh app run` examples. It instructs running `curl -fsSL https://cli.inference.sh | sh` and `infsh login`, which execute remote shell scripts and perform remote authentication, posing execution and network risks.

This skill provides guidance to create consistent character reference sheets and LoRA-based workflows for AI image generation. It instructs remote script download and execution via `curl -fsSL https://cli.inference.sh | sh` and running `infsh` CLI commands that perform network actions.

Dangerous skill: a content-repurposing guide that demonstrates installing and invoking a remote CLI (`curl -fsSL https://cli.inference.sh | sh`) and running networked commands like `infsh login` and `infsh app run`. It shows example workflows to generate images, audio, and posts while instructing execution of piped remote installers and external API calls.

Dangerous skill: instructs downloading and executing a remote installer via `curl -fsSL https://cli.inference.sh | sh` and running `infsh login`. It also directs networked `infsh app run` calls (e.g., `infsh app run tavily/search-assistant`) and writes `results-chart.png`.

This skill provides Chat UI building blocks for React/Next.js. It includes executable install commands like `npx shadcn@latest add https://ui.inference.sh/r/chat.json` and `npx skills add inference-sh/agent-skills@agent-ui`, which perform network fetches and run CLI tooling.

Dangerous skill: contains explicit instructions to run a piped remote installer `curl -fsSL https://cli.inference.sh | sh` and to `infsh login`/`infsh app run` external AI services. The skill is for `book-cover-design` and references local files like `path/to/best-concept.png` and `path/to/final-cover.png`.

Dangerous skill: contains instructions to download and execute a remote installer and to run CLI commands that contact remote services. It instructs `curl -fsSL https://cli.inference.sh | sh && infsh login` and multiple `infsh app run ...` commands that perform network access and local execution.

Dangerous skill installs and runs remote scripts via `https://cli.inference.sh` (includes `curl -fsSL https://cli.inference.sh | sh`). It documents Dia TTS usage with `infsh` commands (e.g., `infsh login`, `infsh app run falai/dia-tts`) and references files like `talking-head.mp4`.

Dangerous skill: includes a piped remote install `curl -fsSL https://cli.inference.sh | sh` and uses `infsh login`, enabling remote shell execution and third-party authentication. It instructs network calls to `https://cli.inference.sh` and to run `infsh app run` against external apps.

This skill documents using the `inference.sh` CLI to run Google Veo video-generation apps. It instructs executing a remote installer via `curl -fsSL https://cli.inference.sh | sh` and running `infsh login`, which executes shell code and triggers networked authentication.

High-risk skill that installs and executes a remote installer via `curl -fsSL https://cli.inference.sh | sh`. It also instructs `infsh login` and running `infsh app run` with external image URLs (e.g., `https://low-res-image.jpg`) and writes `image.json`.

High-risk skill: downloads and executes a remote installer via `curl -fsSL https://cli.inference.sh | sh` and instructs the `infsh` CLI, which makes network calls and uses `INFSH_API_KEY`. It documents saving inputs/results to files like `input.json` and `result.json`.

This skill is a landing-page design guide that integrates AI visuals via the `infsh` CLI. It instructs downloading and executing `curl -fsSL https://cli.inference.sh | sh` and running `infsh`, which contacts `https://inference.sh`.

Dangerous skill: provides examples for accessing 100+ LLMs via the `infsh` CLI and OpenRouter models. It instructs downloading and executing `https://cli.inference.sh` via `curl -fsSL https://cli.inference.sh | sh` and running `infsh login`, enabling remote network execution and credentialed login.

This skill provides newsletter curation guidance and includes concrete `infsh` CLI examples for sourcing and composing issues. It also instructs executing `curl -fsSL https://cli.inference.sh | sh` and running `infsh app run ...` commands, which invoke remote downloads and networked CLI actions.

Provides slide-by-slide guidance and CLI examples for generating investor pitch deck visuals. Contains a remote download-and-execute command (`curl -fsSL https://cli.inference.sh | sh`) and multiple `infsh app run ...` examples plus `infsh login` that run code and perform network operations.

This skill provides Product Hunt launch guidance and CLI commands for generating images and research via `infsh`/`inference.sh`. It instructs running `curl -fsSL https://cli.inference.sh | sh` and multiple `infsh app run` commands, which download and execute remote scripts and perform external network requests.

Dangerous skill: includes a piped shell install `curl -fsSL https://cli.inference.sh | sh` that downloads-and-executes remote code and examples invoking external services via `infsh app run`. It also instructs using credentials (`infsh login`) and switching to `Bearer tokens`.

Discover and install related skills from inference.sh skill registry. Helps find complementary skills for your AI workflow. Use for: skill discovery, workflow expansion, capability exploration. Triggers: related skills, find skills, skill discovery, complementary skills, expand workflow, more capabilities, similar skills, skill suggestions

High-risk skill: installs and executes remote code via `curl -fsSL https://cli.inference.sh | sh`. It also directs `infsh login` and many `infsh app run ...` calls (`openrouter/claude-sonnet-45`, `falai/flux-dev`, `google/veo-3-1-fast`), requiring network access and credentials.

High-risk skill: it instructs running a remote installer (`curl -fsSL https://cli.inference.sh | sh`) and to authenticate via `infsh login`, then call remote apps with `infsh app run`. The content provides CLI examples and prompts for AI product photography and e-commerce image generation.

Dangerous skill: Python SDK documentation for `inference.sh` that shows how to run apps, build agents, manage files, and use webhooks. It contains an `eval(call.args['expression'])` example enabling arbitrary code execution, shows reading `os.environ['INFERENCE_API_KEY']`, and calls webhooks like `https://hooks.slack.com/...`.

High-risk skill for SEO content briefs that instructs installing the `inference.sh` CLI and running `infsh` apps. It runs a piped remote install (`curl -fsSL https://cli.inference.sh | sh`) and issues `infsh login` and `infsh app run` commands, enabling shell execution and outbound network activity.

This skill transcribes audio using the `inference.sh` CLI and supplies `infsh` command examples. It instructs executing a remote installer via `curl -fsSL https://cli.inference.sh | sh` and running `infsh login`, which runs remote code and contacts `inference.sh` services.

Dangerous skill. Generates film and video storyboards and instructs installing and using the `infsh` CLI. It directs execution of a piped remote installer `curl -fsSL https://cli.inference.sh | sh`, runs `infsh` commands that contact `https://inference.sh`, and invokes `infsh login`.

Tool lifecycle UI components for React/Next.js from ui.inference.sh. Display tool calls: pending, progress, approval required, results. Capabilities: tool status, progress indicators, approval flows, results display. Use for: showing agent tool calls, human-in-the-loop approvals, tool output. Triggers: tool ui, tool calls, tool status, tool approval, tool results, agent tools, mcp tools ui, function calling ui, tool lifecycle, tool pending

This skill provides declarative UI widgets for React/Next.js. It includes example CLI commands such as `npx shadcn@latest add https://ui.inference.sh/r/widgets.json` and links to `https://ui.inference.sh`, which demonstrate network access and CLI execution.

This skill provides platform-specific video-ad guidance and CLI-driven generation workflows using `inference.sh`/`infsh`. It instructs executing `curl -fsSL https://cli.inference.sh | sh`, running `infsh app run` commands, and calling `infsh login`, enabling remote downloads, network calls, and credential use.

Dangerous skill: downloads and executes a remote installer via `curl -fsSL https://cli.inference.sh | sh` and runs `infsh` image-generation commands. It also instructs `infsh login` and `infsh app run` which contact `https://inference.sh` (network and credential use risk).

High-risk skill: generates social-media carousels but requires running remote installers and CLI commands. It instructs `curl -fsSL https://cli.inference.sh | sh` and uses `infsh` to call `https://inference.sh` services.

Dangerous skill: downloads and executes a remote installer via `curl -fsSL https://cli.inference.sh | sh`, enabling arbitrary shell code execution. It is a technical blog-writing guide with CLI examples for `infsh` image and benchmark generation.

High-risk skill that installs and runs a remote shell script and makes network calls to `https://cli.inference.sh` and `https://inference.sh`. Provides `infsh` CLI examples for TTS and uses sample files like `input.json` and `speech.json`.

High-risk skill: installs and executes a remote installer via `curl -fsSL https://cli.inference.sh | sh` and invokes the `infsh` CLI. The skill instructs authenticating with `infsh login` and running `infsh app run` commands that contact `https://inference.sh` and operate on local files like `portrait.png`.

High-risk skill that automates Twitter/X via the `infsh` CLI. It instructs running a remote installer (`curl -fsSL https://cli.inference.sh | sh && infsh login`) and executing `infsh app run` commands, which perform network calls and require credentials.

This skill is a Twitter/X thread writing guide that shows how to install and use the `infsh` CLI. It includes a piped installer `curl -fsSL https://cli.inference.sh | sh && infsh login` (remote code execution) and `infsh app run` examples that perform network requests.

Dangerous skill: provides video prompt guidance but instructs running `curl -fsSL https://cli.inference.sh | sh` and `infsh` CLI commands. Those commands enable remote script execution and outbound network calls to `https://cli.inference.sh`.

High-risk skill: provides web search and extraction via the `inference.sh` CLI but instructs running a remote installer with `curl -fsSL https://cli.inference.sh | sh`, which downloads and executes code locally. It also directs networked `infsh app run` calls and an `infsh login` authentication step.