LeadMarina

Schedule a recurring search

create_automation

Create an automation: re-run a search on a schedule and send the results somewhere automatically. Pass cities as plain strings (e.g. ['Austin, TX', 'Dallas, TX']) — they're resolved for you. IT SPENDS LEADS ON EVERY RUN, one per result, so confirm the niche, cities and cadence with the user before creating one. A destination of 'close', 'ghl' or 'google_sheets' requires that integration to be connected already (Close with no saved field mapping sends Close's own fields); 'csv' | 'xlsx' | 'json' emails a file and needs nothing set up. Every city must resolve and the destination must be one of the listed values, or nothing is created and the error says what to fix. Up to 20 automations, 25 cities each. Automations search US places only for now: a city outside the US is refused. Advanced search's options (a country other than the US — a territory included — max_leads, exact_category) are refused here too.

Input Schema

Name	Required	Description
`cities`	Yes	Cities to search, e.g. ['Austin, TX'] (max 25).
`cadence`	Yes	How often it repeats.
`keyword`	Yes	The niche to search, e.g. 'roofing contractor' — up to 100 characters.
`start_at`	No	ISO timestamp for the first run. Defaults to now (runs on the next tick, within 15 min).
`destination`	No	Where each run's leads go. 'none' just saves them to the library.
`spreadsheet_id`	No	google_sheets only: add each run to this existing LeadMarina-created sheet, updating rows already in it.

Output Schema

Name	Required	Description
`id`	Yes
`note`	No
`cities`	No
`cadence`	No
`keyword`	No
`destination`	No	The destination as saved — what each run will actually use.
`next_run_at`	No
`data_sources`	No	Where the sources, licences and notices are for emails and owner names that may come from open datasets and public records. Pass it on with any of this data you share.

Schema Changelog

Changes observed during successful MCP inspections.

Changed2 schema fields changedNov 1, 2026

changedInput schema / properties / keyword / description

Previous value: -"The niche to search, e.g. 'roofing contractor'."New value: +"The niche to search, e.g. 'roofing contractor' — up to 100 characters."

addedInput schema / properties / keyword / maxLength
```
Added value: +100
```

Changed1 schema field changedOct 8, 2026

addedOutput schema / properties / data_sources

Added value: +{
+  "description": "Where the sources, licences and notices are for emails and owner names that may come from open datasets and public records. Pass it on with any of this data you share.",
+  "type": "string"
+}

Changed2 schema fields changedOct 6, 2026

changedInput schema / properties / spreadsheet_id / description

Previous value: -"google_sheets only: append to this existing LeadMarina-created sheet."New value: +"google_sheets only: add each run to this existing LeadMarina-created sheet, updating rows already in it."

addedOutput schema / properties / destination / description

Added value: +"The destination as saved — what each run will actually use."

AddedAug 5, 2026

TDQS

A4.6/5.0

Behavior5/5

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

Annotations already declare this is a non-read-only, non-idempotent, open-world write, but the description adds the crucial cost behavior the annotations cannot express: it spends one lead per result on every run, and the user should be consulted first. It also discloses hard limits (20 automations, 25 cities each), integration prerequisites per destination, and the atomic failure mode ('nothing is created and the error says what to fix').

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

Conciseness4/5

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

The core action and its purpose are front-loaded, followed by the cost warning, then prerequisites and limits. It is dense and a few clauses are long (the Advanced-search exclusion list in particular), but essentially every sentence carries operational information.

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?

For a 6-parameter creation tool that spends credits, this covers cost, quotas, prerequisites, failure behavior, and geographic scope, while an output schema handles return values and annotations handle the safety profile. Nothing an agent needs before calling it is missing.

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

Parameters4/5

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

Schema coverage is 100%, so the schema baseline is 3. The description still adds real meaning: cities are plain strings that get resolved server-side, destination values are grouped by setup requirement (close/ghl/google_sheets need a connected integration; csv/xlsx/json just email a file), and Close falls back to its own fields when no mapping is saved. It says nothing extra about start_at, leaving that to the schema.

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?

The description opens with a specific verb+resource pair ('Create an automation') and immediately defines what an automation is: a scheduled re-run of a search that delivers results automatically. That definition cleanly separates it from list_automations, update_automation, and delete_automation without needing the schema.

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

Usage Guidelines4/5

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

It gives an explicit precondition ('confirm the niche, cities and cadence with the user before creating one') and states when the tool is not usable: US places only, and Advanced-search options like non-US countries, max_leads, or exact_category are refused. It does not explicitly compare itself to update_automation for modifying an existing automation, so it is clear context rather than full routing guidance.

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

All LeadMarina tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.