TextKit

Recommend message

recommend_message

Read-onlyIdempotent

Which text should we send? Ranked SMS copy for a goal (abandoned_cart, welcome, winback, failed_payment, trial, promo, ... or any) by expected revenue per message, each with its evidence: the factors that moved its score, the closest published SMS-only case studies with source links, and how much comes from this account's own sends (clicks, STOPs, revenue from tracked events). Pass your own drafts in candidates to score them against the generated ones. Only uses the offer, code and deadline you give it. Sends and saves nothing.

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": {
      "industry": "<industry>",
      "company_type": "<company_type>",
      "brand": "<brand>",
      "product": "<product>"
    },
    "name": "recommend_message"
  }
}

Input Schema

Name	Required	Description	Default
`code`	No	Optional promo code
`goal`	No	The message type to write, or `any` to compare the types that fit `audience`	any
`link`	No	URL or merge field; default {{link}}
`brand`	Yes	Business name as customers know it
`limit`	No	How many messages to return (default 5, max 20)
`offer`	No	Optional, e.g. "50% off your first month". Without one no candidate has an offer
`product`	Yes	What is being sold, as the customer would say it, e.g. "your second number"
`audience`	No	Who gets it, in words, e.g. "opened the paywall and left"
`deadline`	No	Optional, e.g. "tonight"
`industry`	Yes	The sender's industry
`candidates`	No	Your own drafts to score: a JSON array of up to 10 strings
`first_name`	No	Your contacts have first names, so candidates may use {{first_name}}
`use_history`	No	false ranks on the case studies only
`company_type`	Yes	Kind of business
`first_message`	No	The first text these people get from the number: the brand and opt-out line are added and counted

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedOct 30, 2026
- addedInput schema / properties / limit / description
```
Added value: +"How many messages to return (default 5, max 20)"
```
AddedOct 29, 2026

TDQS

A4.1/5.0

Behavior4/5

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

Annotations already declare readOnly/idempotent/non-destructive, and the description reinforces this with 'Sends and saves nothing' rather than contradicting it. It adds genuinely non-schema behavior: ranking is revenue-per-message based, it uses only the offer/code/deadline supplied, and it discloses the evidence payload (score-moving factors, published case studies with source links, account send history). It does not discuss determinism or latency, so short of a 5.

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?

Front-loaded with the core question, then the ranking basis, then the evidence returned, then the two constraints. Dense but every clause carries information; the parenthetical goal list is slightly redundant with the enum but harmless.

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

Completeness4/5

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

With 15 parameters and no output schema, the description carries the return-value burden and does so (ranked copy plus factors, case studies, and account-derived signal). The 'sends and saves nothing' and 'only uses what you give it' constraints cover the main behavioral risks. Missing only notes on what happens with thin account history beyond the `use_history` flag.

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

Parameters3/5

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

Schema description coverage is 100% and enum/default/min-max details are already in the schema, so the baseline is 3. The description adds marginal but real meaning: that `candidates` are scored against generated copy and that no offer is invented when `offer` is omitted. Most parameters (industry, company_type, audience, first_message) get no extra contextualization in the prose.

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 specific question and names the resource precisely: ranked SMS copy by expected revenue per message, scoped to a goal. It also states the output composition (factors, case studies with links, own-account evidence), so the agent knows exactly what comes back versus other copy/template tools like create_template or send_message.

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?

Clearly establishes this as a pre-send decision tool: 'Which text should we send?' plus 'Sends and saves nothing', which implicitly routes the agent to send_message/create_template for actual sending. It also gives a concrete usage pattern ('Pass your own drafts in candidates to score them against the generated ones'). It does not name a sibling alternative explicitly, so not a 5.

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

All TextKit 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.