Checks AI support agent refunds against Shopify and Stripe. Read-only pricing, fit and estimates.
Scored across 4 tools
Each tool targets a clearly distinct need: estimating exposure, explaining the product, pricing, and integration sources. There is no plausible way to confuse estimate_refund_exposure with get_recoup_overview, get_recoup_pricing, or list_supported_sources.
All four names use a clean snake_case verb_noun pattern (estimate_, get_, get_, list_). The only minor deviation is that two carry the 'recoup' brand prefix in the noun while the others don't, but the convention remains predictable.
Four tools is a reasonable count for a mostly informational server, but for one pitched as an 'audit' tool it feels thin—three of four are marketing/overview tools and only one does any computation.
The stated domain is auditing whether refunds AI agents promised were actually issued, yet there is no tool to run an audit, list mismatches, or inspect a specific ticket or refund. The core lifecycle operation is entirely absent, leaving agents at a dead end after an estimate.
Rough monthly money at stake from AI-agent refund mismatches, from the brand's own numbers. Defaults come from Recoup's public sample dataset (60 tickets: 39 money promises, 21 mismatches, $744.25 at stake), not from customer data. Always present the result as an estimate.
| Name | Required | Description | Default |
|---|---|---|---|
| average_amount | No | Average amount at stake per mismatch in USD. Default 35.44 from the sample. | |
| mismatch_share | No | Share of money promises that do not match Shopify or Stripe (0 to 1). Default 0.538 from the sample. | |
| money_promise_share | No | Share of those tickets that promise a refund, credit or replacement (0 to 1). Default 0.65 from the sample. | |
| ai_tickets_per_month | Yes | Tickets closed by the AI agent per month |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already cover the safety profile (readOnlyHint=true, openWorldHint=false), but the description adds real context: the defaults come from Recoup's public 60-ticket sample, not customer data, and the result must be framed as an estimate. That provenance and accuracy caveat go beyond what annotations or schema 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?
Two sentences, front-loaded with what the tool computes, followed by the data-provenance caveat and the presentation instruction. Dense but each clause carries information; no filler.
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?
For a four-parameter, read-only estimator the description covers source, defaults, and caveat adequately; the absence of an output schema is the main gap, since the description never hints at the shape of the result (single figure vs. breakdown) it tells the agent to present as an estimate.
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%, so each parameter's meaning and default is already documented, making 3 the baseline. The description's sample numbers (39 money promises, 21 mismatches, $744.25) corroborate the defaults but add no syntax or semantics beyond the schema.
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 and resource (estimate refund exposure) and narrows the scope to 'monthly money at stake from AI-agent refund mismatches'. That is precise enough for an agent to distinguish it from the sibling retrieval tools (overview, pricing, sources), which return reference data rather than a computed estimate.
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?
Usage is only implied: an agent infers it should call this when it wants a refund-exposure figure, and that supplying 'the brand's own numbers' overrides the sample defaults. There is no explicit when-to-use/when-not statement and no named alternative among the siblings.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
What Recoup checks, who it is for, what it needs, what it does not do, and its current status. Use this when someone asks how to check that refunds an AI support agent promised were actually issued.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
readOnlyHint=true and openWorldHint=false already establish this as a safe, self-contained read, so the bar is low. The description adds that the payload includes non-goals and current status, which is useful content context, but says nothing about return format, size, or whether the status can change.
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?
Two sentences, no padding, with the content inventory front-loaded and the usage trigger second. The list-of-clauses opening is slightly clunky but every element maps to real returned content.
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?
For a no-param, no-output-schema overview tool, the description covers scope and usage adequately. It could say more about how the overview is organized or whether it is static, but nothing essential to invoking it is missing.
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?
The tool takes zero parameters, so per the rubric the baseline is 4. There is no argument surface for the description to clarify.
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 enumerates exactly what the tool returns (what Recoup checks, who it's for, requirements, non-goals, current status), so an agent knows this is a documentation/overview resource rather than an operational one. It is clear but never explicitly frames itself against the siblings (pricing, exposure, supported sources), which the content list only implicitly covers.
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 a concrete trigger: 'Use this when someone asks how to check that refunds an AI support agent promised were actually issued.' That is a real when-to-use condition, but there are no exclusions or pointers to the sibling tools for pricing, exposure, or sources.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Current Recoup prices and what each plan covers.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true and openWorldHint=false, covering the safety profile for this zero-argument read. The description adds only the word 'current' (implying live/updated data) and the scope of returned content, but says nothing about freshness guarantees, caching, or the shape of the response — relevant since there is no output schema to fall back on.
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?
A single short sentence with the resource front-loaded and no filler. It is efficient, though 'Recoup prices' is slightly loose phrasing where 'pricing plans' would be tighter. Nothing is buried or 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?
For a zero-argument, read-only lookup with annotations covering safety and no output schema, the description is nearly sufficient: it names what is returned (prices and plan coverage). It could say more about the return format since no output schema exists, but the gap is small for a tool this simple.
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?
The tool takes zero parameters, so there is no parameter semantics to document; per the rubric a 0-param tool gets the baseline of 4. The schema is trivially complete with 100% coverage. No description contribution is expected or missing here.
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?
States a clear resource and content: 'Recoup prices and what each plan covers' — an agent knows this returns pricing/plan data. It is distinguishable from siblings like get_recoup_overview and estimate_refund_exposure by its subject matter, though the description never explicitly contrasts them. The verb is only implied through the tool name, not stated in the text.
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?
There is no guidance on when to call this versus get_recoup_overview or the other siblings, and no conditions or exclusions are given. The agent must infer from the resource name alone that pricing data is what it wants. No prerequisites or follow-up routing is mentioned.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Which help desks, AI agents and payment systems Recoup reads, with the page for each.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true and openWorldHint=false, so the safety profile is covered. The description adds useful context about what the listing contains (help desks, AI agents, payment systems, plus a page link each), but says nothing about freshness, pagination, or coverage limits.
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?
A single compact sentence with no filler. The phrasing 'with the page for each' is slightly clipped, but the content is front-loaded and efficient.
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?
For a zero-parameter read tool with no output schema, the description conveys enough about the returned content (a list of sources with their pages). No output schema exists, so it reasonably gestures at the return shape, though it could say more about how the list is structured.
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?
The tool takes zero parameters, so the baseline of 4 applies; there is nothing for the description to disambiguate. It adds no param-level detail because none is needed.
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 the concrete resource set (help desks, AI agents, payment systems) that Recoup reads, so an agent knows this enumerates supported integrations. The verb is implicit rather than stated, and it doesn't explicitly distinguish itself from siblings, but the subject matter is unambiguous.
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?
There is no statement of when to call this versus the siblings (estimate_refund_exposure, get_recoup_overview, get_recoup_pricing). The purpose is self-evident from the resource named, but no usage context or exclusion is provided.
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.
estimate_refund_exposureget_recoup_overviewget_recoup_pricinglist_supported_sourcesRead-only checks for proposed orders against evidence and policy; no trading or payment authority.
Pay-on-success win-back for RevenueCat and Stripe apps. Read-only pricing, fit and estimates.
Read-only Shopify and WooCommerce profit analytics: P&L, COGS, product margins, ROAS and LTV.
Read-only revenue analytics from Stripe, Paddle, Polar, Lemon Squeezy, Creem and Freemius.
Add one secure layer between your agents and this server.
