Synthex Forge — x402 Paid Agent Services

Evaluate Edge Replication

evaluate_edge_replication

Read-onlyIdempotent

Whether a claimed edge survives replication on your own returns. use it rather than verify_edge_replication for edge replication specifically; an object with 7 fields including accepted_fields, detail. $3 USDC.

Input Schema

Name	Required	Description
`claimed`	No	Optional: {sharpe, cagr, max_dd} — your stated figures. When supplied they are compared against the computed ones and a mismatch is a named finding.
`returns`	No	The return series, as a JSON list of numbers. EITHER this or 'returns_csv_b64' is required, never both. At most 200000 observations; a longer series is a 400 naming the limit. NaN and infinity are rejected — they are not a return series.
`frequency`	No	'daily' or 'per_trade'. REQUIRED. A daily series is annualised with sqrt(252); a per_trade series is reported PER TRADE and is never annualised, because a per-trade series has no knowable trades-per-year.
`strategy_shape`	No	Optional: {type, instrument, timeframe}. 'type' is one of trend \| mean_reversion \| breakout \| other and selects the calibration bucket; 'instrument' selects the instrument-class bucket. Instruments are NORMALISED (EUR/USD == EUR_USD == eurusd).
`returns_csv_b64`	No	The return series as a base64-encoded CSV, one number per row. The alternative to 'returns'. Decoded size is capped at 1 MB.
`trials_attempted`	No	How many configurations you tried before keeping this one. REQUIRED, and must be >= 2. This is the number that makes the Deflated Sharpe meaningful: a Sharpe of 2.0 found after 500 tries is priced for what it is. We never assume trials = 1.

Output Schema

Name	Required	Description
`note`	No	A note on how the answer was produced.
`error`	No	The machine-readable error code, when the operation failed.
`detail`	Yes	Human-readable explanation of the verdict.
`verdict`	Yes	The decision this tool returned.
`x402Version`	No	The `x402Version` field of the response.
`accepted_fields`	No	The `accepted_fields` field of the response.
`required_fields`	No	The `required_fields` field of the response.

Schema Changelog

Changes observed during successful MCP inspections.

Changed9 schema fields changedOct 31, 2026

addedOutput schema / properties / accepted_fields

Added value: +{
+  "description": "The `accepted_fields` field of the response.",
+  "type": "object"
+}

addedOutput schema / properties / detail

Added value: +{
+  "description": "Human-readable explanation of the verdict.",
+  "type": "string"
+}

addedOutput schema / properties / error

Added value: +{
+  "description": "The machine-readable error code, when the operation failed.",
+  "type": "string"
+}

addedOutput schema / properties / note

Added value: +{
+  "description": "A note on how the answer was produced.",
+  "type": "string"
+}

addedOutput schema / properties / required_fields

Added value: +{
+  "description": "The `required_fields` field of the response.",
+  "type": "object"
+}

removedOutput schema / properties / status

Removed value: -{
-  "type": "string"
-}

addedOutput schema / properties / verdict

Added value: +{
+  "description": "The decision this tool returned.",
+  "type": "string"
+}

addedOutput schema / properties / x402Version

Added value: +{
+  "description": "The `x402Version` field of the response.",
+  "type": "object"
+}

addedOutput schema / required

Added value: +[
+  "detail",
+  "verdict"
+]

AddedOct 31, 2026

TDQS

A3.6/5.0

Behavior4/5

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

Annotations already cover the safety profile (readOnly, idempotent, non-destructive, closed-world), so the description only needed to add operational context — and it does, disclosing the price ($3 USDC) and the response shape (an object with 7 fields including accepted_fields and detail). It stops short of saying anything about determinism, latency, or whether results are cached, which keeps it from 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.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

It is short and the core claim leads, but the structure is visibly degraded: the second clause begins lowercase mid-sentence, and the response-shape and price fragments are spliced on awkwardly. Every element is useful (routing hint, output shape, cost) yet the packaging reads as truncated rather than crafted.

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 an output schema present, the description need not explain return values, and with 100% schema coverage it need not explain parameters — so the remaining burden is purpose, routing, and cost, all of which are present. The gap is the missing statement of the two practically-required inputs and any note on the trials/deflation logic that gives the result its meaning.

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% — every parameter carries a substantive description explaining defaults, limits, mutual exclusivity of `returns`/`returns_csv_b64`, and why `trials_attempted` matters. The prose adds nothing beyond that, so the baseline of 3 applies rather than a penalty.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description states a specific evaluation: whether a claimed edge survives replication using the caller's own return series, and explicitly names the sibling `verify_edge_replication` as the thing it is not. The phrasing is slightly abstract (no concrete verb like 'test') and the tail sentence about the output object muddies the single-purpose statement, but an agent can still tell what the tool computes.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It gives one routing hint — prefer this over `verify_edge_replication` when the question is replication on your own returns — which is real, if circular ('for edge replication specifically' restates the same thing). There is no guidance on prerequisites (you must supply `frequency` and `trials_attempted`, which the schema marks REQUIRED despite the empty required list), no when-not conditions, and no mention of what to use when you don't have a return series.

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

All Synthex Forge — x402 Paid Agent Services 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.