Agent Embassy

Compute History

compute_history

Idempotent

GPU market memory: cheapest and median asks per class, in a signed receipt. limit defaults to 50. Live: compute_spot_prices. Moves: compute_alerts. POST the tool inputs as JSON to paid_url with an x402 v2 PAYMENT-SIGNATURE header. See https://docs.cdp.coinbase.com/x402 Paid tool — $0.25, USDC on Base or Arbitrum One. Paid URL: https://aemb.pro/v1/markets/compute/history.

Input Schema

Name	Required	Description
`gpu`	No	Optional GPU filter, e.g. 'H100 SXM' (max 64 chars). Omit for whole-market history
`limit`	No	Max history points to return; integer 1-200. Defaults to 50
`idempotency_key`	Yes	Idempotency key: 8-128 printable ASCII characters (no spaces), client-generated, one per logical request — a UUID v4 is recommended. Replaying the same key on this tool returns the original receipt — the action is not re-run, and the first request's receipt wins even if you resend different inputs. Published example keys are rejected. Keys are scoped per tool and per payer account: a key never returns another wallet's receipt.

Output Schema

Name	Required	Description
`how`	Yes	How to pay: POST the tool inputs as JSON to paid_url with an x402 v2 PAYMENT-SIGNATURE header
`tool`	Yes	Tool name, e.g. check_url_custom
`asset`	Yes	Settlement asset
`price`	Yes	Exact price, e.g. "$0.25"
`networks`	Yes	EVM CAIP-2 network ids accepted
`paid_url`	Yes	x402-paid HTTPS URL to POST the tool inputs to
`facilitator`	Yes	x402 facilitator
`payment_required`	Yes	Always true: this MCP tool is a payment pointer; the tool itself is monetized via x402.

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedNov 3, 2026

changedInput schema / properties / idempotency_key / description

Previous value: -"Idempotency key: UUID v4, client-generated, one per logical request. Replaying the same key on this tool returns the original receipt — the action is not re-run, and the first request's receipt wins even if you resend different inputs. Published example keys are rejected. Keys are scoped per tool and per payer account: a key never returns another wallet's receipt."New value: +"Idempotency key: 8-128 printable ASCII characters (no spaces), client-generated, one per logical request — a UUID v4 is recommended. Replaying the same key on this tool returns the original receipt — the action is not re-run, and the first request's receipt wins even if you resend different inputs. Published example keys are rejected. Keys are scoped per tool and per payer account: a key never returns another wallet's receipt."

Changed1 schema field changedOct 23, 2026
- changedOutput schema / properties / price / description
```
Previous value: -"Exact price, e.g. \"$0.05\""New value: +"Exact price, e.g. \"$0.25\""
```
Changed1 schema field changedOct 17, 2026
- changedOutput schema / properties / tool / description
```
Previous value: -"Tool name, e.g. ping"New value: +"Tool name, e.g. check_url_custom"
```
Changed1 schema field changedOct 15, 2026
- changedOutput schema / properties / price / description
```
Previous value: -"Exact price, e.g. \"$0.15\""New value: +"Exact price, e.g. \"$0.05\""
```
Changed1 schema field changedOct 15, 2026
- changedOutput schema / properties / tool / description
```
Previous value: -"Tool name, e.g. verified_check"New value: +"Tool name, e.g. ping"
```

Changed2 schema fields changedSep 26, 2026

removedInput schema / properties / agent_wallet

Removed value: -{
-  "description": "Dev mode only: declares the account wallet (string, min 20 chars) when no x402 payment is present. Ignored in live mode — there the wallet is recovered from the verified payment authorization.",
-  "type": "string"
-}

changedInput schema / properties / idempotency_key / description

Previous value: -"Idempotency key: any non-empty string (a UUID is recommended; no format is enforced). Replaying the same key on this tool returns the original receipt — the action is not re-run, and the first request's receipt wins even if you resend different inputs. Keys are scoped per tool: the same key on another tool is unrelated."New value: +"Idempotency key: UUID v4, client-generated, one per logical request. Replaying the same key on this tool returns the original receipt — the action is not re-run, and the first request's receipt wins even if you resend different inputs. Published example keys are rejected. Keys are scoped per tool and per payer account: a key never returns another wallet's receipt."

Changed4 schema fields changedSep 24, 2026

changedInput schema / properties / agent_wallet / description

Previous value: -"dev mode only"New value: +"Dev mode only: declares the account wallet (string, min 20 chars) when no x402 payment is present. Ignored in live mode — there the wallet is recovered from the verified payment authorization."

addedInput schema / properties / gpu / description

Added value: +"Optional GPU filter, e.g. 'H100 SXM' (max 64 chars). Omit for whole-market history"

addedInput schema / properties / idempotency_key / description

Added value: +"Idempotency key: any non-empty string (a UUID is recommended; no format is enforced). Replaying the same key on this tool returns the original receipt — the action is not re-run, and the first request's receipt wins even if you resend different inputs. Keys are scoped per tool: the same key on another tool is unrelated."

addedInput schema / properties / limit / description

Added value: +"Max history points to return; integer 1-200. Defaults to 50"

First observedSep 23, 2026

TDQS

A4.4/5.0

Behavior5/5

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

The annotations only declare it is non-read-only, non-destructive, idempotent and closed-world; the description carries all the operationally critical context the schema cannot: it is a paid call ($0.25 USDC on Base or Arbitrum One), it requires an x402 v2 PAYMENT-SIGNATURE header, inputs must be POSTed as JSON to the paid URL, and it links to docs. This is exactly the kind of monetary/auth behavior an agent must know before invoking, and it is not duplicative of any structured field.

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 payload/scope sentence is correctly front-loaded, followed by alternatives, payment mechanics, and the endpoint, with no filler. The clipped fragment style ("Moves: compute_alerts.") is dense to the point of being slightly cryptic and could cost a beat of parsing, which keeps it off a 5.

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?

An output schema exists, so return values need no prose. Given three near-fully documented parameters, the description still covers the things structure cannot: pricing, chain, x402 header requirement, transport (POST JSON to paid_url), the concrete endpoint, and sibling routing. Nothing an agent needs to invoke this correctly is missing.

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%, so the schema already documents gpu filtering, the 1-200 integer range, the default of 50, and the full idempotency-key contract. The description only restates "limit defaults to 50," adding no format, syntax, or edge-case meaning beyond the schema, so the baseline 3 applies.

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?

"GPU market memory: cheapest and median asks per class, in a signed receipt" names the resource (GPU market history), the specific payload (cheapest/median ask per class), and the artifact form (signed receipt). It then explicitly separates itself from two siblings: "Live: compute_spot_prices. Moves: compute_alerts." An agent can distinguish this from every other tool in the list without opening a 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?

The description routes the agent by contrasting this historical view against the live-price sibling and the price-movement sibling, which is real guidance rather than generic filler. It lacks an explicit conditional ("use when you need a time series rather than a current quote") and says nothing about when-not to call it, so it stops short of a 5.

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

All Agent Embassy 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.

Resources

Need Help?