Run 250+ cloud AI apps for image, video, LLM, search, and Twitter tasks via the inference.sh CLI.
npx skills add https://github.com/Aadi-110i/PEP-PROJECT --skill infsh-cli-aadi-110i
Please help me install this Agent Skill. Skill: infsh-cli Source: https://github.com/Aadi-110i/PEP-PROJECT/tree/main/skills/infsh-cli Command: npx skills add https://github.com/Aadi-110i/PEP-PROJECT --skill infsh-cli-aadi-110i
belt app run <namespace/app> --input <json>, with automatic local file uploads for images, audio, and video inputs. - App Discovery & Samples: Search, filter, and inspect apps with belt app list, belt app get, and generate ready-to-edit input templates via belt app sample. - Async Task Management: Submit long-running jobs with --no-wait and poll results later using belt task get <task-id>. - Use Case: Generate a product video by running belt app run google/veo-3-1-fast --input '{"prompt": "drone over mountains"}', then post it to Twitter with x/post-create in the same workflow. ## Quick Start Install the CLI with the curl command from the docs, run belt login, then ask the AI to generate an image with the prompt "a cat astronaut" using the falai/flux-dev-lora app.High-intent search queries and answers about installing and using this skill.
Run any app with `belt app run <namespace/app-name> --input '<json>'` after installing the CLI and logging in with `belt login`. Use `belt app sample <app> --save input.json` to generate a template input file you can edit before running.
Use `belt app run falai/flux-dev-lora --input '{"prompt": "your prompt"}'` for images or `belt app run google/veo-3-1-fast --input '{"prompt": "your scene"}'` for videos. The CLI returns URLs to the generated files when the task completes.
Yes, the CLI automatically uploads local files when you provide a path instead of a URL in any input field. Absolute paths, relative paths like ./image.png, and home directory paths like ~/Pictures/photo.jpg are all supported.
Set the INFSH_API_KEY environment variable with your API key, which overrides the local config file. For interactive use, run `belt login` to authenticate through a browser, and verify with `belt me`.
Add the `--no-wait` flag to `belt app run` to submit the job and return immediately with a task ID. Check the status and retrieve results later with `belt task get <task-id>` or save output with `--save result.json`.
Invalid input errors mean your JSON payload does not match the app's schema. Run `belt app get <app>` to see required fields, or generate a correct template with `belt app sample <app> --save input.json` and edit it.
padjeiSearches and retrieves existing images from Salesforce CMS and Data 360 via MCP tools.
Miosa-osaGenerate platform-sized ad images from campaign briefs and brand profiles.
Menschlichkeit-OsterreichExecute over 150 AI applications via the inference.sh CLI.
GoogleCloudPlatformGuides Gemini API usage on Vertex AI with the Gen AI SDK across five languages.