Price derivatives, insurance and energy models with exact sensitivities, checked against bumping.
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
Recompute the sensitivities by central bump-and-revalue on the same scenarios; report each gap, whether they agree within tolerance, and the time each method took.
This tool isn't marked read-only, so it may change something on the server or in an account you connected.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http pricinglab https://mypricinglab.com/lab/mcp
Recompute the sensitivities by central bump-and-revalue on the same scenarios; report each gap, whether they agree within tolerance, and the time each method took.
| Argument | Type | Description |
|---|---|---|
| model* | string | |
| inputs | any |
{
"type": "object",
"title": "check_against_bumpingArguments",
"required": [
"model"
],
"properties": {
"model": {
"type": "string",
"title": "Model"
},
"inputs": {
"anyOf": [
{
"type": "object",
"additionalProperties": true
},
{
"type": "null"
}
],
"title": "Inputs",
"default": null
}
}
}Inputs (name, unit, default, range, demo-server cap), outputs and extra functions of one model.
| Argument | Type | Description |
|---|---|---|
| model* | string |
{
"type": "object",
"title": "describe_modelArguments",
"required": [
"model"
],
"properties": {
"model": {
"type": "string",
"title": "Model"
}
}
}List the available models with a one-line summary each.
No arguments.
{
"type": "object",
"title": "list_modelsArguments",
"properties": {}
}Value and all sensitivities from one adjoint pass. `inputs` overrides defaults by name.
| Argument | Type | Description |
|---|---|---|
| model* | string | |
| inputs | any |
{
"type": "object",
"title": "price_modelArguments",
"required": [
"model"
],
"properties": {
"model": {
"type": "string",
"title": "Model"
},
"inputs": {
"anyOf": [
{
"type": "object",
"additionalProperties": true
},
{
"type": "null"
}
],
"title": "Inputs",
"default": null
}
}
}Run a model's extra calculation, e.g. model='variable_annuity', function='nested'.
| Argument | Type | Description |
|---|---|---|
| model* | string | |
| inputs | any | |
| function* | string |
{
"type": "object",
"title": "run_model_functionArguments",
"required": [
"model",
"function"
],
"properties": {
"model": {
"type": "string",
"title": "Model"
},
"inputs": {
"anyOf": [
{
"type": "object",
"additionalProperties": true
},
{
"type": "null"
}
],
"title": "Inputs",
"default": null
},
"function": {
"type": "string",
"title": "Function"
}
}
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 10, 2026.
No readOnlyHint / destructiveHint annotations, so clients can't tell a read from a write without asking.
Fix: Add annotations to each tool, e.g. { readOnlyHint: true } for tools that only read.