Catalyst Datastreams logo

Catalyst Datastreams

Organization
catalystbyzoho
catalyst-datastreams

Catalyst Data Streams — real-time publish/subscribe channels over WebSocket. Publish from a function with publishData / publish_data, issue a token pair with getTokenPair / get_token_pair, subscribe with DataStreamsWebSocket / ZCDataStreamsWebSocket (Node.js, browser, Java, Python), sendAck, live/earliest/resume subscribe types, 48-hour retention, User Roles vs Custom Logic access control. Trigger on 'Data Streams', 'datastreams', 'real-time channel', 'live updates', 'pub/sub', 'WebSocket subscriber', 'token pair', 'wss-id', 'app_user_id is missing', 'Invalid access control source', 'Authentication failed 3000', '@zcatalyst/datastreams', or 'zcatalyst_sdk.datastreams'.

Overview

Publishercatalystbyzoho
Repositoryagent-skills
Skill namecatalyst-datastreams
Stars
70
Forks
17
Bundled files
4
LicenseApache-2.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 4 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by catalystbyzoho on GitHub. Read the source before you install it.

Installation

Install the Catalyst Datastreams AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/catalystbyzoho/agent-skills/tree/main/skills/catalyst-datastreams
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/catalystbyzoho/agent-skills.git /tmp/agent-skills
mkdir -p .claude/skills
cp -r /tmp/agent-skills/skills/catalyst-datastreams .claude/skills/catalyst-datastreams
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Catalyst Datastreams in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Catalyst Datastreams on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Catalyst Datastreams is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

How It Works

  1. Channels and access control are console-only. Create the channel and choose its access control in Console → Cloud Scale → Data Streams; no SDK, REST, or MCP tool creates channels. Ask the user for the channel ID and its access control type, or list channels with getAllChannels() (control_type 0 = User Roles, 1 = Custom Logic). Load references/datastreams-basics.md for concepts, the token rules, subscribe types, acknowledgement, and the REST API.

  2. Pick the token identifier from the access control — it is not interchangeable.

    • User Roles → userId = an app user's ID from Catalyst Authentication. A console admin's user ID is rejected; connectionName fails with "Mandatory parameter app_user_id is missing."
    • Custom Logic → connectionName, and the token must be generated inside the Advanced I/O function selected as the channel's access control. That function is the token endpoint: it runs your check, then calls getTokenPair. Catalyst never calls it on its own. From any other function or REST: "Invalid access control source!". User-ID tokens on these channels are issued but the WebSocket rejects them (3000).
  3. Server code (publish, token pair, channel reads) — load the runtime reference.

    • Node.js / JavaScript → references/datastreams-javascript-basics.md. Initialize with the app: new DataStreams(await zcAuth.init(req)). Advanced I/O handlers get Node's raw req/res: no req.body, res.status(), or res.json().
    • Python → references/datastreams-python-basics.md. Requires zcatalyst-sdk 1.5.0+ (Python 3.10+). app = zcatalyst_sdk.initialize(req=request), then app.datastreams() (plural). Pass get_token_pair a string channel ID and a keyword user_id= / connection_name=; the result's keys are hyphenated (token["wss-id"]). The subscriber does nothing until connect().
    • Java → references/datastreams-java-basics.md. Dependency is com.zoho.catalyst:java-sdk (2.5.0+) from https://maven.zohodl.com; no datastreams artifact exists.
    • publishData, getAllChannels, getChannelDetails and getLiveCount are admin methods: server only, never in browser or mobile code. getTokenPair also exists in the browser SDK, but a Custom Logic token must come from the configured function.
  4. Subscriber code — read the token correctly and always acknowledge. The token pair's keys are hyphenated: pass token['wss-id'] as zuid, plus token.url and token.key. Subscribe only after the open event. Call sendAck() after every message, or the next message is never delivered. Messages carry data (always a string) and streamingId (plus url/method on API-operation messages). To reconnect, release the old session (old token → unsubscribe()), take a fresh token, subscribe('0'), and reload state from your own store; resuming a dropped session with '-2' redelivers missed messages twice and replays the 48-hour retention. Unsubscribe before every deliberate close; the error event with code 1000 that follows your own unsubscribe() is expected. In Node, getLiveCount() returns { status, data: { connections } }. Java: subscribe() throws the checked ZCDataStreamException, and outside a Catalyst function create the client with new ZCDataStreamsWebSocket(config), not ZCDatastream.getInstance().

  5. Verify end to end before reporting success. Open a subscriber, check getLiveCount() shows it, publish a test message, and confirm the subscriber received it. For anything the references do not cover, fetch the page under https://docs.catalyst.zoho.com/en/cloud-scale/help/data-streams/ as Markdown (<page-url>/index.md), and prefer the references where they disagree.

Security Checklist

  • publishData and the channel-read methods are admin operations: run them only in a Catalyst function, never in a browser or mobile app.
  • Protect the token endpoint. Anyone who can call it can subscribe to the channel, so authenticate the caller before issuing a token.
  • On Custom Logic channels, the configured function is the only gatekeeper: put the real validation there before calling getTokenPair.
  • Do not log or return the token key anywhere except to the client that asked for it.

Triggers

Use this skill for: "Data Streams", "Catalyst Data Streams", "datastreams", "real-time channel", "live updates", "push updates to the browser", "pub/sub", "publish data", publishData, getTokenPair, "token pair", "wss-id", DataStreamsWebSocket, ZCDataStreamsWebSocket, ZCDatastream, "WebSocket subscriber", sendAck, "subscribe type", "resume from last message", "live count", getLiveCount, "channel access control", "User Roles channel", "Custom Logic channel", "app_user_id is missing", "Invalid access control source", "Authentication failed 3000", "reconnect", "duplicate messages after reconnect", "DUPLICATE_VALUE", @zcatalyst/datastreams, zcatalyst_sdk, get_token_pair, publish_data, send_ack, or "Data Streams Python".

References

ReferenceLoad when the query is about…
references/datastreams-basics.mdConcepts, channels, access control (User Roles vs Custom Logic), which token identifier to use, token pair shape, subscribe types, acknowledgement, retention, message fields, REST API endpoints and scopes
references/datastreams-javascript-basics.mdNode.js functions (init, publish, token endpoint, Custom Logic function, channel reads), Node and browser subscribers, package versions
references/datastreams-python-basics.mdPython functions and subscribers: zcatalyst-sdk 1.5.0+, app.datastreams(), get_token_pair argument rules, DataStreamsWebSocket with connect() and set_on_* callbacks, deploying Python functions
references/datastreams-java-basics.mdJava functions and subscribers: Maven setup, ZCDatastream, ZCDataStreamsWebSocket, checked exceptions, running outside Catalyst

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Catalyst Datastreams AI skill do?

Catalyst Data Streams — real-time publish/subscribe channels over WebSocket. Publish from a function with publishData / publish_data, issue a token pair with getTokenPair / get_token_pair, subscribe with DataStreamsWebSocket / ZCDataStreamsWebSocket (Node.js, browser, Java, Python), sendAck, live/earliest/resume subscribe types, 48-hour retention, User Roles vs Custom Logic access control. Trigger on 'Data Streams', 'datastreams', 'real-time channel', 'live updates', 'pub/sub', 'WebSocket subscriber', 'token pair', 'wss-id', 'app_user_id is missing', 'Invalid access control source', 'Authen...

Why use Catalyst Datastreams on TypingMind?

Because you install it once and use it with any model. Catalyst Datastreams is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Catalyst Datastreams in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/catalystbyzoho/agent-skills/tree/main/skills/catalyst-datastreams. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Catalyst Datastreams?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Catalyst Datastreams?

As many as you like. As long as a model supports skills, you can use Catalyst Datastreams with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Catalyst Datastreams AI skill free?

Yes. It is published on GitHub by catalystbyzoho under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇