Agent Skill · Frontline

agents-chat

Test a Frontline agent end-to-end from the terminal using frontline agents chat (Overview/Playground channel). Use when the user wants to send a test message, open a test conversation, continue one, or close one. Distinct from Max chat and from reading conversation transcripts.

Provider: Frontline Path in repo: agents-chat/SKILL.md

Skill body

Agent Test Chat

frontline agents chat talks to an agent the same way the in-app Overview / Playground does: it runs the agent and its active or draft flow. Use it to validate flows end-to-end after building or deploying.

This is a test channel, not production LiveChat/WhatsApp traffic.

Prerequisites

Commands

Open a new test conversation

frontline agents chat

Opens a conversation and runs the agent’s Start flow (no user message).

Send a message (new or continued)

# New conversation + first message
frontline agents chat --message "Hello"

# Continue an existing conversation
frontline agents chat --message "My name is Juan" --conversation-id 123
Flag Description
--agent-id <id> Agent ID (defaults to active agent)
--message <text> Message to send
--conversation-id <id> Continue an existing test conversation (requires --message)
--contact-id <id> Link the conversation to a People record

Close a test conversation

frontline agents chat close <conversationId>

Output

Default (JSON, scripting-friendly):

{
  "ok": true,
  "data": {
    "conversation_id": 123,
    "messages": [...]
  }
}

The CLI normalizes all success responses to { ok: true, data: … }. Use jq '.data.conversation_id' and jq '.data.messages'.

With --pretty, stdout is human-readable: conversation_id plus the latest assistant message (no JSON envelope).

vs other chat commands

Command Purpose
frontline agents chat Test an agent — runs active/draft flow (Playground)
max chat / max "…" Chat with the Max AI assistant (not an agent flow)
frontline agents conversations list/get Read production conversation transcripts (no flow execution)

Typical workflow

frontline agents use <agent-id>
frontline agents chat --message "I need help with my order"
# Note conversation_id from output: jq '.data.conversation_id'
frontline agents chat --message "Order #12345" --conversation-id <id>
frontline agents chat close <id>

After editing a flow, re-run frontline agents chat to validate the updated graph before deploying.

Troubleshooting

Error Cause
agent_not_selected No active agent — run frontline agents use <id> or pass --agent-id
missing_message --conversation-id requires --message to continue

See also

Skill frontmatter

allowed-tools: Bash(frontline:*)

Work with this as data

Every skill here is available over the APIs.io API and to AI agents over MCP.

MCP server

One button, every client — Claude, Cursor, VS Code and the rest.

https://apis.io/mcp

Tools for agent skills

4 MCP tools reach this
  • find_skillsBrowse and filter every skill in the catalog.
  • apis_io_searchSTART HERE — APIs, providers and tags for one query, each with its total.
  • resolveTurn a domain, URL or GitHub org into the provider it belongs to.
  • find_cohortsEvery scored population of providers in the catalog.
All 92 tools →

Call it yourself

curl for this page
This skill
curl "https://apis.io/api/v1/skills/agents-chat"
All agent skills
curl "https://apis.io/api/v1/skills?limit=25"

Discovery needs no key. Ratings and market analysis are Pro.

Get an API key

Free tier, no email required.

A second provider on the same verified email joins the account you already have.