Catalyst by Zoho — Skill Index

This is the routing layer. Load the most specific matching skill — do not answer from this file alone.

Philosophy

• Prefer MCP over asking. If CatalystbyZoho_* tools are available, use them. Never ask the user to copy IDs from the console.
• Default to Development. Always target the Development environment unless the user explicitly says "production" or "deploy to prod".
• "Build an app" means Slate + Function by default. When a user says "build an app", "create an app", or "make a simple app" without specifying backend-only, the default output is a Slate frontend + Advanced I/O function backend. Do NOT build only a function and call it an app. If the user's intent is clearly backend-only (e.g. "build an API", "write a function"), skip Slate.
• Prefer Functions over AppSail for simple HTTP. Functions are cheaper (per-invocation billing), simpler to deploy, and require no infrastructure management. Reach for AppSail only when the use case genuinely requires a persistent process, or a custom runtime.
• Show cost before building. For any new infrastructure (functions, AppSail, Stratus buckets), load catalyst-pricing and give a brief estimate before writing code. Most small projects stay within free tier — say so when true.
• Recommend the current service, not the deprecated one. File Store → Stratus. Event Listeners → Signals. Cron → Job Scheduling. Never mention the deprecated name in generated code or config.
• Warn before the region bites. Circuits and Integration Functions do not work in most data centers. Check the user's DC before suggesting them.

How It Works

1. Pre-flight — Check that .catalystrc and catalyst.json exist. If missing, use MCP to get org/project IDs and run catalyst init --org <orgId> -p <projectId> -ni. Never use interactive catalyst init.
2. MCP check — Look for CatalystbyZoho_* tools. If available, use MCP to fetch org/project IDs instead of asking the user.
3. Route — Match the query to the most specific service in the routing table below.
4. Load lazily — Read ONLY the single reference file needed for the current step. Do NOT preload multiple skills upfront. For "build an app" requests: (a) assume Slate frontend + AIO function unless the user says backend-only, (b) sketch the architecture briefly and confirm with the user before building, (c) then load one reference file per service as you write each part — catalyst-slate for the frontend, catalyst-functions for the backend.
5. Cost check — Only load catalyst-pricing if the user specifically asks about cost, or if the plan includes AppSail, Stratus, or other paid-tier services. Skip for basic Functions + DataStore projects (likely free tier).
6. Answer — Provide code examples using the user's platform (Node.js, Python, Java, Web, or Mobile).

Triggers

Use this skill for queries containing: Catalyst, zcatalyst, AppSail, Data Store, ZCQL, Cache, Stratus, Slate, NoSQL, Zia Services, QuickML, API Gateway, Connections, Zoho MCP, CatalystbyZoho, catalyst init, catalyst deploy, catalyst serve, zcatalyst-sdk-node, catalyst-config.json, Catalyst pricing, "build on Zoho's platform", or "deploy to Catalyst". Do NOT use for generic Zoho CRM questions unless Catalyst is the target.

Pre-flight gate (project-mutating tasks only)

Step 1 — MCP check (do this before anything else for infrastructure tasks): Look for CatalystbyZoho_* tools in your tool list.

• If present — use MCP to fetch org/project IDs. Never ask the user to copy IDs from the console. How to call MCP tools correctly:
  1. ZohoMCP_getSchema takes query_params, NOT body:

     ZohoMCP_getSchema({ query_params: { tool_name: "CatalystbyZoho_List_All_Projects" } })
     

     Copy
  2. Always call ZohoMCP_getSchema first for any CatalystbyZoho_* tool — never guess the argument shape. Many tools require path_variables (e.g. project_id) that are invisible without the schema.
  3. ZohoMCP_executeTool takes a body with this shape:

     ZohoMCP_executeTool({ body: {
       tool_name: "CatalystbyZoho_List_All_Functions",
       arguments: {
         path_variables: { project_id: "31594000000127002" },
         headers: {},
         body: {}
       }
     }})
     

     Copy
  4. Tools with no required path variables (e.g. List_All_Organizations, List_All_Projects) can be called with arguments: {}.
• If NOT present and the task creates resources, writes files, deploys, or reads project IDs — HARD STOP. Do NOT write any code or create any files. Tell the user:

  "Zoho MCP needs to be connected before I can work with your Catalyst project. Load the catalyst-zoho-mcp skill to set it up — it takes under a minute." Do not proceed until CatalystbyZoho_* tools are visible.

This gate applies only when the task writes Catalyst project files, deploys, reads project/environment IDs, or performs MCP project operations. Skip it for informational questions (pricing, architecture advice, service selection, "what is Catalyst?", install help, SDK usage, or any question that doesn't require an initialized project).

For project-mutating tasks: verify .catalystrc and catalyst.json exist in the working directory.

• If missing → use MCP tools to get the org ID and project ID, then run:

  catalyst init --org <orgId> -p <projectId> -ni
  

  Copy
  Never ask the user to run catalyst init interactively. Never create .catalystrc or catalyst.json yourself — they must be generated by the CLI. Note: NI mode can only link an existing project. If no project exists yet, tell the user to create one in the Catalyst console, then come back — you'll link it with catalyst init -ni.

Catalyst at a Glance

New to Catalyst? Here's what each service does in one line:

Skill Routing Table

Never Use (deprecated + regionally restricted)

Deprecated services — do not use for new projects

Users who signed up after August 27, 2025 cannot access these components at all.

Regionally restricted services — check DC before recommending

These services are unavailable in the listed data centers. Building with them for users in those regions results in runtime failures.

How to check: Ask the user which data center their Catalyst account uses, or look for the DC code in their console URL (e.g., catalyst.zoho.in → IN, catalyst.zoho.eu → EU).

Non-Interactive Mode

CLI v1.27.0+ supports non-interactive (NI) mode — the CLI takes all answers from flags and sensible defaults instead of prompting.

Enable: Add -ni (or --non-interactive) to any supported command, or set ZCATALYST_NON_INTERACTIVE=1 in the environment.

Disabled commands: functions:setup → use functions:add -ni; functions:shell → use functions:execute

Blocked options: --remote on functions:delete and client:delete is not allowed in NI mode.

Tip: catalyst <command> -ni --help shows the NI-specific help for that command.

.catalystrc — Reading org and project context

After catalyst init, a .catalystrc file is written to the project root. Read it at the start of any session to identify the active project without asking the user.

Actual format:

{
  "defaults": { "project": 1, "env": 1 },
  "actives":  { "project": 1, "env": 1 },
  "projects": [
    {
      "idx": 1,
      "id": "<projectId>",
      "name": "<projectName>",
      "domain": { "id": "<domainId>", "name": "<projectName>-<orgId>.development" },
      "env": [{ "idx": 1, "id": "<orgId>", "name": "Development", "type": 3 }]
    }
  ]
}

How to read the active project ID and org ID:

1. Read defaults.project → this is the active project's idx value
2. Find the entry in projects where idx matches → get its id → this is the project ID to use with -p
3. From the same entry, read env[0].id → this is the org ID to use with --org

Quick-reference: Top gotchas

• TERMINOLOGY: always say "organization" or "org", NEVER "portal" — the catalyst init CLI prompt says "Select a default Catalyst portal" but this is the CLI's legacy wording; the correct term is organization
• .catalystrc / catalyst.json missing → run catalyst init --org <orgId> -p <projectId> -ni (get IDs from MCP); never create these files manually, never use interactive catalyst init
• All Catalyst CLI commands default to non-interactive (CLI v1.27.0+) — always use -ni with --org, -p, --name, --type, --stack flags; never fall back to interactive/arrow-key menus in agent context
• catalyst functions:add non-interactive — catalyst functions:add --name <n> --type <type> --stack <stack> -ni
• ZCQL result rows are wrapped → rows.map(r => r.TableName) to unwrap
• ZCQL max 300 rows/query → use LIMIT offset, count for pagination
• ZAID differs between Dev and Prod → #1 auth issue in production
• AppSail PORT → process.env.X_ZOHO_CATALYST_LISTEN_PORT (not PORT)
• Cache values are strings only → serialize/deserialize JSON yourself
• Stratus bucket names are globally unique → use {app-name}-{project-id-prefix}
• Slate → function calls are cross-domain → use full URL + generateAuthToken() + CORS whitelist
• DataStore App User permissions OFF by default → enable in Console → Table → Scopes & Permissions
• catalyst-config.json format → deployment + execution keys only; NOT function or entry_point
• Advanced I/O node20 req → raw http.IncomingMessage; use sendJson(res, ...) helper, not res.json()
• signOut() requires a redirect URL → catalyst.auth.signOut(redirectURL)

Documentation

• Main docs: https://docs.catalyst.zoho.com/en/llms.txt
• Node.js SDK (complete, ~420 KB): https://docs.catalyst.zoho.com/en/sdk/nodejs/v2/llms-full.md
• Web SDK (complete, ~116 KB): https://docs.catalyst.zoho.com/en/sdk/web/v4/llms-full.md
• Python SDK (complete, ~489 KB): https://docs.catalyst.zoho.com/en/sdk/python/v1/llms-full.md
• Java SDK (complete, ~407 KB): https://docs.catalyst.zoho.com/en/sdk/java/v1/llms-full.md

The per-SDK llms-full.md files are complete references — download and grep for the method you need rather than reading them whole. The overview/index.md pages are thin landing stubs; do not rely on them for API details.

• CLI reference: https://docs.catalyst.zoho.com/en/cli/v1/cli-command-reference/index.md
• Pricing: https://catalyst.zoho.com/pricing.html

Fetching docs: llms.txt is the markdown index of every page. Append index.md to any docs page URL to get its markdown instead of HTML. A response starting with <!DOCTYPE html> means the page does not exist.