Hyper CLI

SkillAI & models

Use the Hyper CLI to run Hyper marketing skills from a terminal. Use when the user wants to use `hyperai`, inspect live tool schemas, translate MCP tool names from a skill into CLI commands, switch connected accounts, or run marketing tools outside an MCP-native agent host.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the Hyper CLI skill

What this skill tells your AI

The instructions your AI receives, as published by hyperfx-ai/marketing-skills in skills/hyper-cli/SKILL.md and read by ahel’s review.

Use hyperai to run any marketing skill from a terminal. A skill names each tool once, by its canonical name, for example gmail_messages_send. The CLI runs that name directly.

Requirements

If hyperai info fails, stop and tell the user to authenticate the CLI before calling tools.

The four commands

NeedCommand
Confirm auth and workspacehyperai info
Find a toolhyperai search "<what you want to do>"
Read a tool's input schemahyperai describe <name> --parameters
Run a toolhyperai call <name> --json '{...}'

<name> is the canonical tool name from the skill, such as meta_ads_adaccount_list. The CLI also accepts the same name as separate words (meta-ads adaccount list); a skill never uses that form.

If your CLI version rejects the canonical name, run hyperai search "<name>": the result prints the exact hyperai call ... line to use. Update the CLI with hyperai update.

Connected accounts

NeedCommand
List the connected accounts for a toolkithyperai call connections_list --json '{"toolkit_id": "gmail"}'
Switch the active account for this shellhyperai call connections_use --json '{"toolkit_id": "gmail", "auth_id": "<auth_id>"}'

The active account is kept per shell session, the same way the MCP keeps it per session.

Arguments

  • Pass arguments as one JSON object with --json. The CLI validates them against the tool's schema before calling.
  • When a call fails with an argument error, run hyperai describe <name> --parameters and read the schema. Do not retry the same arguments.
  • Large results: add --raw to get the unformatted JSON.

Example

hyperai info
hyperai search "send an email"
hyperai describe gmail_messages_send --parameters
hyperai call gmail_messages_send --json '{"to": "jane@example.com", "subject": "Hello", "body": "Hi Jane"}'

Out of scope

NeedSkill
What to send, to whom, in which sequencethe marketing skill that owns the job, for example cold-email-outreach
MCP client setupthe Hyper MCP page at https://app.hyperfx.ai/mcp

Signals

GitHub stars
86
Forks
15
Last commit
Sep 2026
Advanced
Catalog kind
skill
Gateway key
hyper-cli
Source
github.com/hyperfx-ai/marketing-skills