Skip to main content
Glama
Sign In
Arsel MCP Server

Arsel MCP Server

Official
by Arsel-SA
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Hybrid

Create Automation

create-automation

Build a draft automation graph from steps and connections, returning any missing config issues to fix before the user enables it in the dashboard.

Instructions

Create an automation as a draft: a journey each contact who matches the trigger goes through. Send steps and connections to build it, or only name for an empty draft. This server cannot enable or disable an automation: once issues is empty, the user turns it on from the Arsel dashboard. Saving steps messages nobody.

Shape. An automation is a graph. steps holds every step, the trigger included (key trigger), and connections link them by key. A step leaves by default, except a condition (condition_met, condition_not_met) and a wait_for_event (event_received, timeout). A branch with no connection ends the run. Every step must be reachable from the trigger, no connection may lead back to an earlier step, and branches that split must rejoin at one step or not at all. At most 100 steps.

Look up what the steps name before writing them. Tags from list-tags, lists from list-lists, segments from list-segments, email templates from list-templates, sending domains from list-email-domains, SMS senders from list-sms-senders, WhatsApp templates from list-whatsapp-templates, custom properties from list-contact-properties, events from list-events, campaigns from the channel's campaign list tool.

Conditions and waits take a filter in the segment rule format. To react to a message this automation sent, use a message_engagement rule with target_type: automation_step and the send step's key as target_id; that step must always run before.

Drafts may be unfinished. The automation comes back with issues, each naming a step_key and the config path still missing or wrong. Fix them with update-automation. A graph that breaks a rule of the shape above is refused and nothing is saved.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "name": "<name>"
    },
    "name": "create-automation"
  }
}

Input Schema

NameRequiredDescriptionDefault
nameYes
stepsNoEvery step, the trigger included. Send it with `connections`; together they replace what was there.
connectionsNoSend it with `steps`.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv1.2.0

TDQS

A5/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations only declare readOnly=false, destructive=false, idempotent=false, openWorld=false. The description adds substantial context beyond them: the result is always a draft, activation is out of scope, invalid graphs are refused with nothing saved, and a saved draft can be incomplete with `issues` naming a `step_key` and config `path`. No contradiction with the destructiveHint=false claim, since a refused graph persists nothing.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loads the core action and its limitation, then uses bold section headers ('Shape', 'Look up...', 'Conditions and waits', 'Drafts may be unfinished') so a long payload stays scannable. Every block carries a constraint an agent would otherwise get wrong; there is no filler or restatement of the name.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

No output schema exists, and the description compensates by explaining the return shape (`issues` with `step_key` and missing `path`) and the failure mode (refused graph, nothing saved). Combined with the graph invariants, lookup dependencies, and the `message_engagement`/`automation_step` constraint, an agent has everything needed to build a valid draft.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Adds graph-level meaning the schema cannot express per-property: `steps` must include the trigger keyed `trigger`, `connections` link by key, exit types per step kind (`default`, `condition_met`/`condition_not_met`, `event_received`/`timeout`), reachability from trigger, no back-edges, branch rejoin rule, and the 100-step cap. It also says `name` alone yields an empty draft and that steps+connections together replace what was there.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Opens with a precise verb+resource+state: 'Create an automation as a `draft`: a journey each contact who matches the trigger goes through.' It also names the sibling it hands off to (update-automation) for fixing issues, so an agent can separate creation from repair without opening either schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

States explicit boundaries: this server cannot enable or disable an automation, the user turns it on from the dashboard, and saving steps messages nobody. It also gives the prerequisite workflow ('Look up what the steps name before writing them') mapped to specific list-* tools per reference type, which is when-to-use guidance at the operational level.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

Other Tools