cogny clisupportfirst 15 calls free

crisp mined from the shell.

List conversations, read message history, pull people profiles and events, and post replies — from Bash over the Crisp MCP. Turn live-chat transcripts into JSON you can mine for objections, FAQs, and ad-copy verbatims.

1. sign up, free for the first calls (uses your git email)npx @cogny/cli init --agent claude-code
2. see every Crisp toolcogny tools list
3. call any tool from bash, no MCP wiringcogny tools call mcp__Crisp__list_conversations --input '{…}'

Start free @cogny/cli on npm

01two paths

you're a person

Reading support history in the Crisp inbox is manual scroll-and-skim. The CLI pulls whole conversations as JSON so an agent can surface recurring objections, content gaps, and the exact phrases customers use. Connect Crisp once in Cogny Solo and let your coding agent, or a scheduled loop, do the querying for you.

Start Solo

you're an agent

Sign yourself up with one command. The CLI issues a key, wires nothing, and exposes every Crisp tool over JSON-RPC. The first 15 calls are free.

npx @cogny/cli init --agent claude-code

02tools call

real Crisp tools.

Tool names as the Crisp MCP serves them. Send --input JSON, get JSON back. The payloads are illustrative.

mcp__Crisp__list_conversations

List a website inbox, newest first.

cogny tools call mcp__Crisp__list_conversations \ --input '{"website_id":"8c842c8e-4f1e-4f9a-9d2a-1a2b3c4d5e6f","page_number":1,"filter_state":"resolved"}'

example response

{
  "conversations": [
    {
      "session_id": "session_a1",
      "state": "resolved",
      "last_message": "Thanks, that fixed it!",
      "meta": {
        "nickname": "Jordan",
        "email": "jordan@example.com"
      }
    },
    {
      "session_id": "session_a2",
      "state": "resolved",
      "last_message": "Does it work with WooCommerce?",
      "meta": {
        "nickname": "Sam",
        "email": "sam@example.com"
      }
    }
  ]
}

mcp__Crisp__get_conversation_messages

Read full message history for one conversation.

cogny tools call mcp__Crisp__get_conversation_messages \ --input '{"website_id":"8c842c8e-4f1e-4f9a-9d2a-1a2b3c4d5e6f","session_id":"session_a2"}'

example response

[
  {
    "from": "user",
    "type": "text",
    "content": "Does it work with WooCommerce?",
    "timestamp": 1718530000000
  },
  {
    "from": "operator",
    "type": "text",
    "content": "Yes — connect your store keys once and you are set.",
    "timestamp": 1718530120000
  }
]

mcp__Crisp__get_website_statistics
Inbox volume and response stats.
- cogny tools call mcp__Crisp__get_website_statistics \ --input '{"website_id":"8c842c8e-4f1e-4f9a-9d2a-1a2b3c4d5e6f"}'
example response
```
{
  "conversations": {
    "total": 1822,
    "resolved": 1690,
    "pending": 14,
    "unresolved": 118
  }
}
```

03tools list

the rest of the surface.

A sample of the other Crisp tools. Run cogny tools list --json for every tool and its input schema.

mcp__Crisp__get_website
mcp__Crisp__list_website_operators
mcp__Crisp__get_conversation
mcp__Crisp__get_conversation_meta
mcp__Crisp__send_message_in_conversation
mcp__Crisp__list_people_profiles
mcp__Crisp__get_people_profile
mcp__Crisp__list_people_conversations
mcp__Crisp__list_people_events

04then a verdict

calls from a script. verdicts from cogny.

Every write is logged with who authorised it. Launches your agent makes (a new campaign, a new ad) become measured tickets automatically; for other changes, your agent offers to register them. Cogny measures each in your own data, three days later for ads and seven for content and SEO, and closes it as worked, partial, didn't work or inconclusive.

worked
Moved the number. Every measured win has a share card.
partial
Some of it did.
didn't work
On the record anyway, so it isn't proposed again.
inconclusive
Not enough data to call.

How the loop works

05questions

questions.

Is this an official Crisp CLI?: No — it is the Cogny CLI calling the Crisp MCP over the cogny-mcp-proxy. You connect your Crisp workspace once in Cogny, then read conversations, profiles, and stats from Bash. Crisp IM SARL is not affiliated.
What is the conversation-mining use case?: Pull resolved conversations with `list_conversations` + `get_conversation_messages` and feed the transcripts to an agent to extract objections, FAQ gaps, and the verbatim language customers use for landing-page and ad copy.
Can it reply, or only read?: Both. `send_message_in_conversation` posts a reply and `update_conversation_state` changes status; the rest of the surface is read tools for mining and reporting.

06related

Building an agent? The full tool map for every server is one fetch: /cli/llms.txt

crisp, then a verdict.

Start free All CLIs