Belt
- 4.6k installs
- 6 repo stars
- Updated July 13, 2026
- belt-sh/cli
belt is the inference.sh cloud CLI for running AI apps, managing agent knowledge, searching skills, and executing MCP connector tools.
About
belt is the inference.sh cloud CLI for AI agents packaged as a single roughly 4MB binary with no runtime dependencies. Install via curl installer, then belt login and belt me for account status. The apps subsystem searches, inspects schemas, generates sample inputs, and runs models such as openai/gpt-image-2, pruna image tools, google/veo-2 video, tavily/search, and elevenlabs/tts with --json and --save flags for piping and media output. Knowledge commands provide semantic search, typed listing, create from files or stdin, and delete across skill, concept, observation, reference, and preference types. Skills support on-demand belt skill use from the store or GitHub, persistent belt skill add with agent auto-detection, listing, and upload publishing. MCP connectors list, search, connect, list tools, and run tool calls like slack send_message. belt suggest unifies search across apps, skills, and knowledge. Tips recommend belt app sample for unfamiliar schemas, --json for structured output, and belt update for latest releases.
- Single ~4MB binary with apps, knowledge, skills, MCP connectors, and unified suggest search.
- belt app run supports image, video, search, and audio models with --input JSON and --save media output.
- Knowledge types include skill, concept, observation, reference, and preference with semantic search.
- Skills run on-demand via belt skill use or install persistently with belt skill add per agent.
- MCP workflow covers belt mcp connect, tools listing, and belt mcp run for connector tool execution.
Belt by the numbers
- 4,587 all-time installs (skills.sh)
- +410 installs in the week ending Aug 5, 2026 (Skillselion tracking)
- Ranked #44 of 550 CLI & Terminal skills by installs in the Skillselion catalog
- Security screen: HIGH risk (skills.sh audit)
- Data as of Aug 5, 2026 (Skillselion catalog sync)
belt capabilities & compatibility
- Capabilities
- search and run 250+ hosted ai apps with json i/o · semantic knowledge create, search, list, and del · on demand and installed skill execution from sto · mcp connector connect, tool listing, and tool ex · unified suggest search across apps, skills, and
- Works with
- slack · openai · anthropic
- Use cases
- orchestration · image generation · web search
- Runs
- Hosted SaaS
- Pricing
- Bring your own API key
What belt says it does
belt is the cloud platform cli for ai agents. single ~4mb binary, no runtime dependencies.
npx skills add https://github.com/belt-sh/cli --skill beltAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 4.6k |
|---|---|
| repo stars | ★ 6 |
| Security audit | 1 / 3 scanners passed |
| Last updated | July 13, 2026 |
| Repository | belt-sh/cli ↗ |
How do agents run hosted AI models, persist knowledge, discover skills, and call MCP tools from one lightweight CLI?
Run 250+ AI apps, manage semantic knowledge, search skills, and connect MCP connectors through the belt cloud CLI binary.
Who is it for?
Developers and agents who need a unified CLI for model apps, semantic memory, skills, and MCP connectors.
Skip if: Skip when the task needs only local code edits without external AI app or connector orchestration.
When should I use this skill?
User mentions belt CLI, inference.sh apps, belt know, belt skill, belt mcp, or belt suggest unified search.
What you get
Installed belt binary with logged-in session, executed app runs, stored knowledge entries, and connected MCP tools via documented commands.
- saved model outputs
- sample input JSON
- MCP connections
By the numbers
- Single ~4MB binary with no runtime dependencies
- Supports 250+ hosted AI apps via belt app commands
Files
belt cli
belt is the cloud platform cli for ai agents. single ~4mb binary, no runtime dependencies.
install
curl -fsSL cli.inference.sh | sh
belt login
belt meapps — run 250+ ai models
belt app search "image" # find apps
belt app get openai/gpt-image-2 # view schema
belt app sample openai/gpt-image-2 --save in.json # generate sample input
belt app run openai/gpt-image-2 --input in.json # run it
belt app run openai/gpt-image-2 --input '{"prompt": "..."}' --save output.pngcommon apps:
- image:
openai/gpt-image-2,reve/create,pruna/p-image - upscale/edit:
pruna/p-image-upscale,pruna/p-image-edit - video:
google/veo-2,seedance/seedance-2-i2v - search:
tavily/search,exa/search - audio:
elevenlabs/tts
knowledge — persistent agent memory
belt know search "query" # semantic search
belt know list --type observation # list by type
belt know get namespace/name # get details
belt know create ./file.md --type concept # create from file
echo "learned X" | belt know create - --name x --type observation # from stdin
belt know delete <id>types: skill, concept, observation, reference, preference
skills — reusable workflows
belt skill search "deployment" # search registry
belt skill store --featured # browse featured
# use on-demand (no install, stdout)
belt skill use namespace/skill-name # from store
belt skill use github.com/user/repo # from github
belt skill use user/repo --skill name # pick from multi-skill repo
# install persistently
belt skill add namespace/skill-name # auto-detects agents
belt skill add ns/name --agent claude-code
belt skill list # list installed
belt skill upload ./my-skill # publishconnectors — mcp servers
belt mcp list # available connectors
belt mcp search "slack" # search
belt mcp connect slack # connect
belt mcp tools slack # list tools
belt mcp run slack send_message --input '{"channel": "#general", "text": "hello"}'suggest — unified search
belt suggest "how to generate images" # searches apps + skills + knowledgetips
- use
--jsonfor structured output when piping - use
--save filenameto save media outputs directly belt app samplegenerates valid input — start there for unfamiliar appsbelt updateto get latest version
Related skills
Lark SharedWire Feishu/Lark Open Platform auth, tokens, and shared API helpers into agent-built bots and internal tools.514k
Runcomfy CliLet their coding agent call any RunComfy model from the terminal or scripts without managing separate SDKs.298k30
Skills CliDiscover, install, update, and manage reusable agent capabilities from the open Skillselion ecosystem directly inside their coding agent.270k71TzstCreate, extract, list, and script `.tzst` / `.tar.zst` archives from the terminal with the right compression, streaming, and conflict flags.267k71XdropPull files from an Xdrop share URL, decrypt them locally, and land outputs in a chosen directory with optional JSON for automation.266k71
WizardAuthor interactive bash setup wizards that open URLs, capture secrets, write .env upserts, and set GitHub Actions secrets step by step.104k202kHow it compares
Choose belt when terminal-first discovery, schema inspection, and MCP wiring matter more than embedding one vendor SDK in application code.
FAQ
How do I start with an unfamiliar app schema?
Run belt app sample to generate valid input JSON before belt app run.
Can skills run without installing them?
Yes. belt skill use runs store or GitHub skills on demand to stdout without persistent install.
What knowledge types does belt support?
skill, concept, observation, reference, and preference entries searchable via belt know search.
Is Belt safe to install?
skills.sh reports 1 of 3 security scanners passed. Review the Security Audits panel on this page before installing in production.