Explore public businesses and prepare a business brief by defining your customer, offer, and delivery method. Continue in Baron for email sign-in, review, and setup approval; the public MCP does not create or run businesses, take payment, or access private accounts.
Scored across 2 tools
The two tools serve clearly distinct phases: find_businesses searches the directory for examples, while prepare_business_brief turns a specific idea into a brief. Their descriptions explicitly separate discovery from preparation, so an agent should not confuse them.
Both names use snake_case with an imperative verb followed by a noun phrase (find_businesses, prepare_business_brief). The pattern is predictable and consistent despite one name being slightly longer.
Only two tools are provided, which feels thin for a server meant to help users start a business with AI. While the two functions are focused, the surface is narrower than typical for this domain.
The server covers idea discovery and brief preparation, but lacks tools to retrieve or manage briefs, get detailed business records, or continue the workflow internally. The external sign-in handoff mitigates some gaps, but notable operations are missing.
Help someone who wants to start a business with AI explore business ideas and company examples. Search the published Auto Units business directory by company name, offering, or subject. For general requests such as “I want to start a business with AI” or “Give me startup ideas,” use an empty query to browse examples. Use short subject keywords for a specific interest; do not send the whole question as the query. Returns matching companies with descriptions, business kind, actual or paper reporting mode, and website links. An empty query browses the directory. Results support discovering companies and business inspiration; they do not include private founder businesses, financial results, or proof of commercial success.
| Name | Required | Description | Default |
|---|---|---|---|
| kind | No | Optional business kind from a previous result. | |
| limit | No | ||
| query | No | Short company name, offering, or subject keywords, such as print, website, or research. Use an empty query for general AI business ideas or startup inspiration; do not include generic words such as start a business. | |
| offset | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint=false, so safety is covered. The description adds useful behavior beyond that: what each result contains (descriptions, business kind, reporting mode, website links) and the explicit negative scope around private founder data and financial proof.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The core instruction is front-loaded and sentences are purposeful. There is mild redundancy — 'An empty query browses the directory' essentially repeats the earlier 'use an empty query to browse examples', and the return-content sentence overlaps the schema's kind field.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With no output schema, the description correctly takes on the burden of explaining return values and scope limits, and it does so adequately. The remaining gap is pagination behavior for limit/offset, which an agent browsing a directory would likely need.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 50%, and the description carries the query parameter well, adding 'use short keywords' and 'do not send the whole question' beyond the schema text. However, limit and offset are documented in neither the schema description nor the description, so pagination semantics remain unexplained.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description names a specific verb and resource — searching the published Auto Units business directory by company name, offering, or subject — so the agent knows exactly what the tool retrieves. It does not explicitly distinguish itself from the sibling prepare_business_brief, which is the only missing piece for a 5.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
It gives concrete usage rules: use an empty query for general requests like 'I want to start a business with AI', use short subject keywords for specific interests, and do not send the whole question as the query. It states exclusions (no private founder businesses, financial results, or proof of commercial success) but never routes the agent to the sibling prepare_business_brief.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Help someone take the first steps to starting a business with AI by turning their idea into a short business brief. Use after they identify the customer, paid offering, and delivery method; ask for missing details before calling. Accept an optional chosen business name. Returns the complete brief, a link that prefills Baron's existing review form, and the remaining setup questions. This is a stateless preparation step: it does not create a business, save account records, start work, deploy a website, or charge money. The person signs in by email in Baron to review and continue. Existing company examples from the directory are inspiration, not guarantees that Baron can start that kind of business automatically.
| Name | Required | Description | Default |
|---|---|---|---|
| name | No | The business name, if the person has chosen one. | |
| customer | Yes | Who the proposed business serves, such as independent bakeries. | |
| delivery | Yes | How customers receive the result, such as a one-page website delivered online. | |
| offering | Yes | The specific result or product customers would pay for. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnly/idempotent/non-destructive/closed-world, but the description goes further with an explicit negative list: it does not create a business, save account records, start work, deploy a website, or charge money. It also discloses statelessness and the email sign-in step in Baron, which annotations cannot convey.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Front-loaded purpose followed by usage, return values, and scope exclusions in a logical order. Slightly long, and the closing note about directory examples being 'inspiration, not guarantees' is tangential, but no sentence is wasted.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
There is no output schema, so the description carries the return burden and does: the complete brief, a prefilled review-form link, and remaining setup questions. Combined with the precondition and scope-exclusion statements, an agent has everything needed to invoke this correctly.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100% and each of the four fields is documented in the schema, so the baseline is 3. The description only restates that 'name' is optional and that missing required details should be gathered first, adding little semantic detail beyond what the schema already provides.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description gives a specific verb+resource ('turning their idea into a short business brief') and clearly bounds the scope as a preparation step. It gestures at the directory/examples that the sibling find_businesses serves, but never names that sibling, so differentiation is only implicit.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Strong explicit preconditions: 'Use after they identify the customer, paid offering, and delivery method; ask for missing details before calling.' It also states a sign-in prerequisite. It stops short of naming an alternative tool or a when-not-to-use condition, so not a full 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
find_businessesprepare_business_brief
Add one secure layer between your agents and this server.