Market evidence with receipts: every claim resolves to a real stored record you can fetch back.
Scored across 5 tools
Each tool has a distinct role: category_warmth orients on held data, research_product triggers a background harvest, search_evidence reads the corpus, get_receipt verifies ids, and compare_formats runs a specific analysis. The trigger/read split between research_product and search_evidence is clearly explained in the descriptions, though the three evidence-related tools sit close enough to warrant care.
All names are snake_case and four follow a verb_noun pattern (compare_formats, get_receipt, research_product, search_evidence). category_warmth is a noun phrase rather than verb_noun, a minor deviation in an otherwise consistent set.
Five tools is well-scoped for an evidence-research server, covering orientation, harvesting, search, verification, and one analysis. No tool feels redundant or missing at the count level.
The surface covers the core lifecycle: orient on holdings, start research, search results, verify receipts, and compare formats. Minor gaps exist (e.g. no explicit way to inspect or cancel an in-flight harvest), but agents can work around these via polling search_evidence.
How much is already held for a category, and therefore whether asking about it is instant and free or slow and expensive. Check this BEFORE starting a report. With no category it lists what is held, so orient here instead of guessing slugs.
| Name | Required | Description | Default |
|---|---|---|---|
| category | No | The category, for example "running shoes". Omit to list every held category. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare read-only, idempotent, and non-destructive behavior. The description adds useful context beyond that: the cost profile of querying a category and the fact that omitting category lists all held categories, which is not visible from annotations alone.
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 description is compact and front-loaded: it states the core concept first, then the command-like usage guidance, then a practical orientation tip. Every sentence contributes, with no filler or repetition of the schema.
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 simple one-parameter read-only tool, the description covers purpose, timing, and both invocation modes. The exact output shape is not described, but there is no output schema and the annotations already cover the safety profile, so nothing critical 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 input schema already describes the single optional category parameter with 100% coverage, including the omission behavior. The description reinforces that omission behavior but does not add meaningful new semantic detail, so the baseline score of 3 is appropriate.
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 defines category warmth as how much is already held for a category and explains the practical consequence (instant and free vs. slow and expensive). It clearly frames the tool as a pre-report check, which makes its purpose distinct from siblings like search_evidence or get_receipt.
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 explicitly says to check before starting a report and explains the no-category usage mode as a way to orient rather than guess slugs. It does not name alternative sibling tools or give when-not-to-use exclusions, but the context is clear.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Video versus static advertising for a category, computed from how long real ads actually ran rather than from opinion. Returns a verdict, its confidence, and the ads behind it.
| Name | Required | Description | Default |
|---|---|---|---|
| category | Yes | The product category. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already provide readOnlyHint, idempotentHint, and destructiveHint, so safety is covered. The description adds meaningful behavioral context by stating the methodology (based on real ad run lengths, not opinion) and disclosing what is returned: a verdict, confidence, and supporting ads.
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 deliver the core purpose, the distinguishing methodology, and the output composition with no redundant wording. The most important information is front-loaded.
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 simple read-only, idempotent tool with one fully documented parameter, the description is complete. It explains what the tool computes, why it is reliable, and what the return will contain, which is especially helpful given there is no output schema.
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%, with the single 'category' parameter already described as 'The product category.' The description only repeats that the comparison is 'for a category' and does not add value such as allowed formats, examples, or edge cases. Baseline 3 is appropriate given full schema coverage.
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 clearly identifies the tool's focus: comparing video versus static advertising for a category and returning a verdict. It cites a specific resource (ad formats) and the underlying evidence, though it does not explicitly differentiate itself from sibling tools by name.
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?
The description implies the tool is appropriate when an empirical, data-driven comparison is needed, since it is 'computed from how long real ads actually ran rather than from opinion.' However, it does not explicitly say when to use this tool over siblings like category_warmth or search_evidence, nor does it mention any conditions when it should not be used.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Resolve receipt ids to the real records behind them. This is how you CHECK a claim: if an id does not resolve, the claim citing it was not real. Takes one id or many. Long records are truncated with notice; pass full: true for complete text.
| Name | Required | Description | Default |
|---|---|---|---|
| full | No | Return complete record text with no truncation. Default false. | |
| receiptIds | Yes | Receipt ids, as returned by any other tool. Up to 50. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnly, idempotent, and non-destructive behavior. The description adds meaningful behavioral context beyond that: records are truncated with notice, full:true returns complete text, and an unresolved id means the cited claim was not real. This gives the agent useful expectations not encoded in the schema or annotations.
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?
Three sentences with no wasted words: purpose, use case, and parameter caveat are each front-loaded and clearly separated. The description earns every sentence and is easy to parse quickly.
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 simple read-only lookup tool with two well-documented parameters and no output schema, the description covers purpose, usage, behavior, and the one nuanced parameter (full). Nothing essential is missing for an agent to select and invoke the tool 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 coverage is 100%, so the structured fields already document the parameters. The description adds value by explaining that receiptIds can be one or many and that full:true disables truncation, which clarifies the practical effect of the parameters beyond their schema definitions.
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: it resolves receipt ids to the real records behind them, and immediately gives the key use case—checking whether a claim is real. This distinguishes it from the sibling tools, which are about searching and comparing rather than resolving a concrete id.
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?
The description clearly states when to use this tool: to CHECK a claim by seeing if an id resolves. It does not explicitly name alternatives or state when not to use it, but the claim-checking context is specific enough that an agent will not confuse it with category_warmth, compare_formats, or search_evidence.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Start collecting evidence on a product or topic from public sources. RETURNS AT ONCE: the harvest runs in the background, the first records land within a minute, and a full run takes minutes. Then call search_evidence or category_warmth for the same subject after about 30 seconds and keep calling while the counts grow; stop when they stop. Every receipt returned along the way is already a real record. Calling this twice for one subject joins the same run rather than starting another.
| Name | Required | Description | Default |
|---|---|---|---|
| terms | No | What to ask about: single concepts like "sizing", "durability". Optional. | |
| subject | Yes | A product name, a topic, or a product URL. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations only declare readOnlyHint=false, openWorldHint=true, idempotentHint=true, destructiveHint=false; the description adds the operationally critical facts the agent cannot infer — that it returns immediately, that records begin landing within a minute, that a full run takes minutes, and that returned receipts are already real records.
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 return behavior is front-loaded with 'RETURNS AT ONCE', and each subsequent sentence carries distinct information: background execution, polling cadence and stop condition, receipt validity, and idempotent re-calls. No sentence restates the schema or annotations.
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 and only two fully documented parameters, the description covers the remaining unknowns an agent needs: that results are not returned synchronously, how to observe progress, when to stop, and that duplicate calls are safe. Nothing required to invoke it correctly 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?
Schema description coverage is 100% (subject = product name, topic, or URL; terms = optional concepts), so the schema already carries the parameter meaning. The description reinforces 'subject' and 'product or topic' but adds no syntax, format, or terms guidance beyond the schema, so the baseline of 3 applies.
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 states a specific verb and resource ('Start collecting evidence on a product or topic from public sources') and immediately identifies its role as the starter of an asynchronous harvest, which is exactly what separates it from retrieval siblings like search_evidence and category_warmth.
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 an explicit follow-up procedure: poll search_evidence or category_warmth after ~30 seconds, keep calling while counts grow, stop when they stop. It also resolves the obvious ambiguity about repeat invocation by stating that a second call for the same subject joins the existing run rather than starting a new one.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search the retained corpus for what people actually said about a product or topic. Returns how many independent records exist, across how many channels, whether that clears the corroboration threshold, and the loudest few quotes with their receipt ids. Costs nothing and touches no network.
| Name | Required | Description | Default |
|---|---|---|---|
| after | No | Optional. ISO date (2026-01-31): only records their authors wrote on or after it. Undated records sit inside no window. | |
| query | Yes | What to look for: "sizing", "battery life", "runs small". | |
| before | No | Optional. ISO date: only records written on or before it. | |
| phrase | No | Optional. Match the words as one ordered phrase, no any word fallback. Build phrases from content words. | |
| sources | No | Optional. Only these sources (reddit, appstore, sec-edgar...). | |
| category | No | Optional. Narrow to one product category. | |
| minChannels | No | Optional. Demand at least this many distinct channels for a finding verdict. Can only demote. | |
| sourceClasses | No | Optional. Filter by what kind of speaker: consumer_voice (reddit, reviews...), practitioner (github...), institutional (regulators, filings). | |
| excludeSources | No | Optional. Drop these sources, e.g. ["sec-edgar", "cpsc"] to drop the institutional records from a consumer question. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, idempotentHint=true, destructiveHint=false, and the description adds valuable behavioral context: it explicitly states the tool 'Costs nothing and touches no network,' and describes what it returns (counts, channels, threshold, quotes with receipt IDs). This goes beyond the structured annotations by detailing the output scope and operational characteristics.
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 description is a single, dense sentence that front-loads the core purpose ('Search the retained corpus') and packs in the return details and operational notes without fluff. Every clause adds value, and the structure is logical — purpose, then result summary, then cost/network note.
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 tool with 9 parameters but no output schema, the description covers the key return elements (counts, channels, threshold, quotes with receipt IDs) and notes that it's free and offline. It doesn't explain the corroboration threshold in detail, but the schema handles parameter specifics. The description is sufficiently complete for an agent to understand what to expect when calling the tool.
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 all parameters are documented in the schema itself. The description does not add extra meaning to individual parameters; it refers to the general search behavior but not to specific parameters. Given the full schema coverage, the baseline of 3 is appropriate.
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 starts with a specific verb and resource: 'Search the retained corpus for what people actually said about a product or topic.' It clearly states the tool's action and scope, and distinguishes it from siblings by describing its unique return (counts, channels, threshold, quotes with receipt ids) — something none of the sibling tools do.
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?
The description implies when to use the tool (to search for evidence from the retained corpus) but does not explicitly compare to alternatives or state when not to use it. No exclusions or 'use instead' guidance is provided, leaving the agent to infer context from the tool's name and purpose.
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.
research_productcategory_warmthcompare_formatsget_receiptsearch_evidenceCryptographically anchored evidence for agents: verified run receipts, proof-gated settlement.
Check claims against a fact-store: consistent, contradicts, or unverifiable — with a receipt.
Ed25519-signed verdicts resolved against real ground truth. Free claim/verify; paid bonded dispute.
Evidence-backed crypto due diligence with sources, freshness, and a runtime receipt on every call.
Add one secure layer between your agents and this server.
