io.github.Kromatic-Innovation/fermi is an MCP server published by io.github.Kromatic-Innovation, hosted at fermi.kromatic.com. When last checked on 2026-11-21 it was working: it answers and lists its tools. It declares 7 tools, including estimate_quantity_or_probability, get_estimate_result, list_my_estimates, record_actual, send_feedback.
Calibrated probability ranges for forecasting questions, with the assumptions behind them.
5 connections through 5 shared values. Tap any node.
Squares are this server's values; dots are other servers sharing them. A shared address or publisher is a lead, not proof of a common operator.
Inferred from the host name and headers: what a server runs on, not who runs it.
Estimates an uncertain number, or the probability that an event happens by a date, from a question in plain language, and returns a calibrated range and probability distribution. Use this when the user asks how much, how many, how big or how long something is or will be (market size, cost, count, price, duration), or asks a yes/no question about the future: "will X happen by <date>?", "what are the odds / chances / likelihood of X?". For a yes/no question, set unit to "probability (0-1)": point_estimate is then Fermi's best probability that the event happens (0.15 means 15%), and estimation_interval is how far that probability could plausibly be off (e.g. 0.05 to 0.35); the beta distribution describes the same uncertainty. There is no separate yes/no verdict. Returns point_estimate, estimation_interval (low/high: the range that should contain the true value, or for a yes/no question the true probability, with probability coverage_probability, default 90%), the fitted distribution (distribution_family and distribution_parameters), reasoning_summary, assumptions, estimate_id, and next_steps.simulate. To combine several estimates into a model with a time axis and get percentiles, NPV or payback out, feed them into neumann (https://neumann.kromatic.com) as simulation root nodes -- every successful estimate carries a next_steps.simulate block with the arguments already prepared. (neumann is Kromatic's companion Monte Carlo simulation server; fermi itself does not simulate.) The neumann URLs above are production. A non-production fermi still emits next_steps.simulate against its own configured neumann base URL, so read that block rather than these links whenever you are not on the production host. Tiers: 1 = fast, answers in the same call, no web search, and the only tier available without an account. 2 = grounded, answers in the same call, searches the web. 3 = deep research, runs in the background: the call returns status "pending" and an estimate_id; read the result later with get_estimate_result. Not for: running a simulation or combining estimates (use neumann); reading prediction-market prices or placing bets; financial, medical or legal advice. Tiers 2 and 3, and tier 1 once the free estimates are used up, need an account. If the user wants one of these and is not signed in, call the tool as normal. If the user's app supports sign-in, it will ask the user to sign in, and you can call the tool again after that. Do not send the user to their connector or app settings to sign in first. Only if the user's app cannot sign in (a REST caller, or an agent with no connector), create an account with POST https://fermi.kromatic.com/api/v1/accounts (it returns an api_key once; a new account starts with 3 credits) and pass that key as the api_key argument. Cost: without an account, 3 free tier-1 estimates per UTC day; once they are used up, calls are refused with a sign-in requirement until 00:00 UTC. With an account, tier 1/2/3 costs 1/3/10 credits ($0.50/$1.50/$5.00 at $0.50 per credit). Tiers 1 and 2 charge only if the estimate succeeds, and a failed call does not use up a free estimate; tier 3 charges when it starts and refunds if it fails. Every call is a new estimate and is charged again, even for the same question. `credits_charged` and `billing` in the result report what was spent.
| Parameter | Type | Description |
|---|---|---|
| unit * | string | Unit of the answer, e.g. "USD", "people", "hotel rooms", "days". For a yes/no event question use "probability (0-1)": the answer is then a probability between 0 and 1 (0.15 = 15%). |
| model | any | Optional model id for that provider. It must be the model configured for that tier and provider, so normally omit it. |
| api_key | any | Your Fermi API key. Omit it if your connector is already signed in, or to use Fermi without an account. Get one with POST https://fermi.kromatic.com/api/v1/accounts (returned once). |
| context | any | Anything the estimate should take into account: recent facts or news, and the user's own numbers or definitions. The server supplies today's date. |
| provider | any | Optional AI provider for the estimate, "openai" or "anthropic". Omit to use the default. Tiers 2 and 3 accept only "openai". |
| question * | string | The question in plain language, one quantity or one event per call. Quantity: "How many hotel rooms will Lisbon have in 2030?". Yes/no event: "Will any EU country make a four-day work week law before 2030-01-01?". |
| callback_url | any | Optional URL that receives the finished or failed result as a signed POST, so a server need not keep asking. Chat clients usually cannot receive it: read the result with get_estimate_result instead. |
| source_inputs | any | Optional documents or data for the estimate to use. Each item has source_type ("qualitative_text", "spreadsheet" or "structured_file") and either content (text) or structured_data (an object, or a list of row objects); title, file_name and media_type are optional. |
{
"name": "estimate_quantity_or_probability",
"title": "Estimate a number, or the probability an event happens by a date",
"description": "Estimates an uncertain number, or the probability that an event happens by a date, from a question in plain language, and returns a calibrated range and probability distribution. Use this when the user asks how much, how many, how big or how long something is or will be (market size, cost, count, price, duration), or asks a yes/no question about the future: \"will X happen by <date>?\", \"what are the odds / chances / likelihood of X?\".\n\nFor a yes/no question, set unit to \"probability (0-1)\": point_estimate is then Fermi's best probability that the event happens (0.15 means 15%), and estimation_interval is how far that probability could plausibly be off (e.g. 0.05 to 0.35); the beta distribution describes the same uncertainty. There is no separate yes/no verdict.\n\nReturns point_estimate, estimation_interval (low/high: the range that should contain the true value, or for a yes/no question the true probability, with probability coverage_probability, default 90%), the fitted distribution (distribution_family and distribution_parameters), reasoning_summary, assumptions, estimate_id, and next_steps.simulate. To combine several estimates into a model with a time axis and get percentiles, NPV or payback out, feed them into neumann (https://neumann.kromatic.com) as simulation root nodes -- every successful estimate carries a next_steps.simulate block with the arguments already prepared. (neumann is Kromatic's companion Monte Carlo simulation server; fermi itself does not simulate.) The neumann URLs above are production. A non-production fermi still emits next_steps.simulate against its own configured neumann base URL, so read that block rather than these links whenever you are not on the production host.\n\nTiers: 1 = fast, answers in the same call, no web search, and the only tier available without an account. 2 = grounded, answers in the same call, searches the web. 3 = deep research, runs in the background: the call returns status \"pending\" and an estimate_id; read the result later with get_estimate_result.\n\nNot for: running a simulation or combining estimates (use neumann); reading prediction-market prices or placing bets; financial, medical or legal advice.\n\nTiers 2 and 3, and tier 1 once the free estimates are used up, need an account. If the user wants one of these and is not signed in, call the tool as normal. If the user's app supports sign-in, it will ask the user to sign in, and you can call the tool again after that. Do not send the user to their connector or app settings to sign in first. Only if the user's app cannot sign in (a REST caller, or an agent with no connector), create an account with POST https://fermi.kromatic.com/api/v1/accounts (it returns an api_key once; a new account starts with 3 credits) and pass that key as the api_key argument.\n\nCost: without an account, 3 free tier-1 estimates per UTC day; once they are used up, calls are refused with a sign-in requirement until 00:00 UTC. With an account, tier 1/2/3 costs 1/3/10 credits ($0.50/$1.50/$5.00 at $0.50 per credit). Tiers 1 and 2 charge only if the estimate succeeds, and a failed call does not use up a free estimate; tier 3 charges when it starts and refunds if it fails. Every call is a new estimate and is charged again, even for the same question. `credits_charged` and `billing` in the result report what was spent.",
"inputSchema": {
"type": "object",
"title": "estimate_quantity_or_probabilityArguments",
"required": [
"question",
"unit",
"target_horizon",
"disclaimer_acknowledged"
],
"properties": {
"unit": {
"type": "string",
"title": "Unit",
"description": "Unit of the answer, e.g. \"USD\", \"people\", \"hotel rooms\", \"days\". For a yes/no event question use \"probability (0-1)\": the answer is then a probability between 0 and 1 (0.15 = 15%)."
},
"model": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Model",
"default": null,
"description": "Optional model id for that provider. It must be the model configured for that tier and provider, so normally omit it."
},
"api_key": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Api Key",
"default": null,
"description": "Your Fermi API key. Omit it if your connector is already signed in, or to use Fermi without an account. Get one with POST https://fermi.kromatic.com/api/v1/accounts (returned once)."
},
"context": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Context",
"default": null,
"description": "Anything the estimate should take into account: recent facts or news, and the user's own numbers or definitions. The server supplies today's date."
},
"provider": {
"anyOf": [
{
"enum": [
"openai",
"anthropic"
],
"type": "string"
},
{
"type": "null"
}
],
"title": "Provider",
"default": null,
"description": "Optional AI provider for the estimate, \"openai\" or \"anthropic\". Omit to use the default. Tiers 2 and 3 accept only \"openai\"."
},
"question": {
"type": "string",
"title": "Question",
"description": "The question in plain language, one quantity or one event per call. Quantity: \"How many hotel rooms will Lisbon have in 2030?\". Yes/no event: \"Will any EU country make a four-day work week law before 2030-01-01?\"."
},
"callback_url": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Callback Url",
"default": null,
"description": "Optional URL that receives the finished or failed result as a signed POST, so a server need not keep asking. Chat clients usually cannot receive it: read the result with get_estimate_result instead."
},
"source_inputs": {
"anyOf": [
{
"type": "array",
"items": {
"type": "object",
"additionalProperties": true
}
},
{
"type": "null"
}
],
"title": "Source Inputs",
"default": null,
"description": "Optional documents or data for the estimate to use. Each item has source_type (\"qualitative_text\", \"spreadsheet\" or \"structured_file\") and either content (text) or structured_data (an object, or a list of row objects); title, file_name and media_type are optional."
},
"target_horizon": {
"type": "string",
"title": "Target Horizon",
"description": "When the answer applies. Prefer an absolute date or year: \"by 2027-06-30\", \"2030\", \"end of 2028\", or \"current\" for a value today. Relative phrases like \"in the next 12 months\" are read from the server's current UTC date."
},
"requested_tier_id": {
"type": "integer",
"title": "Requested Tier Id",
"default": 1,
"description": "1 = fast: answers in the same call, no web search; the only tier without an account. 2 = grounded: answers in the same call, searches the web; account only. 3 = deep research: runs in the background and searches the web; returns status \"pending\" and an estimate_id to read later with get_estimate_result; account only. Default 1. Without an account, tiers 2 and 3 are refused with a sign-in requirement; nothing is downgraded or charged."
},
"coverage_probability": {
"type": "number",
"title": "Coverage Probability",
"default": 0.9,
"description": "How likely the returned interval is to contain the true value, strictly between 0 and 1. Default 0.9, a 90% interval. Use 0.95 for a \"95% interval\", 0.8 for an \"80% range\"."
},
"disclaimer_acknowledged": {
"type": "boolean",
"title": "Disclaimer Acknowledged",
"description": "Must be true. Confirms the result is a probabilistic estimate for decision support, not financial, medical or legal advice. A call with false is refused."
}
}
},
"outputSchema": {
"type": "object",
"title": "EstimateQuantityOrProbabilityResult",
"properties": {
"model": {
"type": "string",
"description": "Model that produced it."
},
"status": {
"enum": [
"pending",
"completed",
"failed"
],
"type": "string",
"description": "Only on results that are not finished yet or that come from get_estimate_result: \"pending\" (tier 3 still running), \"completed\" or \"failed\"."
},
"billing": {
"type": "object",
"properties": {
"principal": {
"enum": [
"anonymous",
"account"
],
"type": "string",
"description": "\"anonymous\" (no account) or \"account\"."
},
"account_id": {
"type": [
"string",
"null"
],
"description": "Account id; null without an account."
},
"charged_to": {
"enum": [
"anonymous_quota",
"credits",
"none"
],
"type": "string",
"description": "What this call drew on: the free daily allowance, account credits, or \"none\" (free calls)."
},
"auth_source": {
"enum": [
"bearer",
"api_key",
"anonymous"
],
"type": "string",
"description": "How you were recognised: \"bearer\" (signed-in connector or token), \"api_key\" (the argument) or \"anonymous\"."
},
"is_unlimited": {
"type": "boolean",
"description": "True for an internal account that is never charged."
},
"account_status": {
"type": [
"string",
"null"
],
"description": "\"created\" if this call created the account, otherwise \"matched\"; null without an account."
},
"credit_balance": {
"type": [
"integer",
"null"
],
"description": "Credits left on the account; null without an account."
},
"credits_charged": {
"type": "integer",
"description": "Credits this call spent."
},
"anonymous_quota_resets_at": {
"type": [
"string",
"null"
],
"description": "When the free allowance resets (next 00:00 UTC, ISO 8601)."
},
"anonymous_estimates_remaining_today": {
"type": [
"integer",
"null"
],
"description": "Free tier-1 estimates left today without an account; null for an account."
}
},
"description": "Who this call was billed to and what is left. Present on every Fermi result.",
"additionalProperties": true
},
"tier_id": {
"type": "integer",
"description": "The tier that produced it (1, 2 or 3)."
},
"provider": {
"type": "string",
"description": "AI provider that produced it."
},
"range_low": {
"type": "number",
"description": "Same as estimation_interval.low."
},
"created_at": {
"type": "string",
"description": "When it was made (ISO 8601)."
},
"disclaimer": {
"type": "string",
"description": "Not financial advice notice."
},
"next_steps": {
"type": [
"object",
"null"
],
"description": "Ready-made input for neumann, Kromatic's Monte Carlo simulation server. simulate.root_node (distribution_type and distribution_params) can be added to a neumann model as one uncertain input; when the family has no direct neumann match, simulate.solve_args are the arguments for neumann's convert_range_to_distribution instead. simulate.note says which, and simulate.mcp_url is the neumann server to call.",
"additionalProperties": true
},
"range_high": {
"type": "number",
"description": "Same as estimation_interval.high."
},
"request_id": {
"type": "string",
"description": "Id of this request, for support."
},
"assumptions": {
"type": "array",
"items": {
"type": "string"
},
"description": "The assumptions the estimate rests on."
},
"estimate_id": {
"type": "string",
"description": "Id of this estimate, for get_estimate_result and list_my_estimates."
},
"error_detail": {
"type": [
"string",
"null"
],
"description": "When failed: what went wrong."
},
"point_estimate": {
"type": "number",
"description": "The single best value. For a yes/no question asked with unit \"probability (0-1)\", this is the probability that the event happens (0.15 = 15%)."
},
"credits_charged": {
"type": "integer",
"description": "Credits this estimate cost (0 without an account)."
},
"confidence_level": {
"enum": [
"low",
"medium",
"high"
],
"type": "string",
"description": "Set only from coverage_probability: \"high\" for 0.95 or more, \"medium\" for 0.9 up to 0.95, \"low\" below 0.9. It is not a judgement of the estimate's quality."
},
"credits_refunded": {
"type": [
"integer",
"null"
],
"description": "When failed: credits given back."
},
"research_sources": {
"type": [
"array",
"null"
],
"items": {
"type": "object",
"additionalProperties": true
},
"description": "Tiers 2 and 3: web sources used (url, title, snippet, relevance_note)."
},
"reasoning_summary": {
"type": "string",
"description": "How the estimate was reached, in a few sentences."
},
"verification_flags": {
"type": [
"array",
"null"
],
"items": {
"type": "string"
},
"description": "Tier 3: issues the second check raised."
},
"distribution_family": {
"type": "string",
"description": "Shape of the fitted probability distribution: normal, log_normal (positive and right-skewed), uniform, constant, beta (values between 0 and 1, used for probabilities and rates), triangle, gamma, exponential or poisson (counts)."
},
"estimation_interval": {
"type": "object",
"properties": {
"low": {
"type": "number",
"description": "Lower end."
},
"high": {
"type": "number",
"description": "Upper end."
}
},
"description": "The range that should contain the true value with probability coverage_probability. For a probability question, both ends are probabilities between 0 and 1.",
"additionalProperties": true
},
"retry_after_seconds": {
"type": [
"integer",
"null"
],
"description": "When pending: seconds to wait before reading it again."
},
"coverage_probability": {
"type": "number",
"description": "Chance the interval contains the true value (0.9 = 90%)."
},
"verification_summary": {
"type": [
"string",
"null"
],
"description": "Tier 3: what a second check found."
},
"distribution_parameters": {
"type": "object",
"description": "Parameters of that distribution by name, e.g. alpha and beta for beta, mu and sigma for log_normal.",
"additionalProperties": {
"type": "number"
}
},
"decomposition_suggestions": {
"type": [
"array",
"null"
],
"items": {
"type": "object",
"additionalProperties": true
},
"description": "Sub-questions (sub_question, reasoning) worth estimating separately to sharpen this one. Each would be a new estimate call at the usual cost."
}
},
"description": "A finished estimate (tiers 1 and 2, and tier 3 once done), or for tier 3 a status \"pending\" placeholder with estimate_id and retry_after_seconds.",
"additionalProperties": true
},
"annotations": {
"title": "Estimate a number, or the probability an event happens by a date",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}Reads one estimate you already requested, by its estimate_id. Use this when a tier-3 (deep research) estimate came back with status "pending" and you want to know whether it is finished, or to show an earlier estimate again. Tier-3 results are only delivered this way (or to a callback_url, which chat clients usually cannot receive). Returns status: "pending" (with retry_after_seconds: how long to wait before asking again), "completed" (with the same fields a finished estimate returns), or "failed" (with error_detail; any credits taken are refunded). You can only read estimates made with the same credentials. Without an account you are recognised by IP address and browser or client, so a different network or client may not find the estimate. To find an id you lost, use list_my_estimates. Not for: making a new estimate. Cost: free. Reading a result never spends anything.
| Parameter | Type | Description |
|---|---|---|
| api_key | any | The same api_key used when the estimate was requested (omit it if none was used, or if your connector is signed in). Estimates are visible only to the credentials that made them. |
| estimate_id * | string | The estimate_id returned by estimate_quantity_or_probability or listed by list_my_estimates. |
{
"name": "get_estimate_result",
"title": "Get the result of an estimate you already requested",
"description": "Reads one estimate you already requested, by its estimate_id. Use this when a tier-3 (deep research) estimate came back with status \"pending\" and you want to know whether it is finished, or to show an earlier estimate again. Tier-3 results are only delivered this way (or to a callback_url, which chat clients usually cannot receive).\n\nReturns status: \"pending\" (with retry_after_seconds: how long to wait before asking again), \"completed\" (with the same fields a finished estimate returns), or \"failed\" (with error_detail; any credits taken are refunded).\n\nYou can only read estimates made with the same credentials. Without an account you are recognised by IP address and browser or client, so a different network or client may not find the estimate. To find an id you lost, use list_my_estimates. Not for: making a new estimate.\n\nCost: free. Reading a result never spends anything.",
"inputSchema": {
"type": "object",
"title": "get_estimate_resultArguments",
"required": [
"estimate_id"
],
"properties": {
"api_key": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Api Key",
"default": null,
"description": "The same api_key used when the estimate was requested (omit it if none was used, or if your connector is signed in). Estimates are visible only to the credentials that made them."
},
"estimate_id": {
"type": "string",
"title": "Estimate Id",
"description": "The estimate_id returned by estimate_quantity_or_probability or listed by list_my_estimates."
}
}
},
"outputSchema": {
"type": "object",
"title": "GetEstimateResultResult",
"properties": {
"model": {
"type": [
"string",
"null"
],
"description": "Model that produced it."
},
"status": {
"enum": [
"pending",
"completed",
"failed",
null
],
"type": [
"string",
"null"
],
"description": "Only on results that are not finished yet or that come from get_estimate_result: \"pending\" (tier 3 still running), \"completed\" or \"failed\"."
},
"billing": {
"type": "object",
"properties": {
"principal": {
"enum": [
"anonymous",
"account"
],
"type": "string",
"description": "\"anonymous\" (no account) or \"account\"."
},
"account_id": {
"type": [
"string",
"null"
],
"description": "Account id; null without an account."
},
"charged_to": {
"enum": [
"anonymous_quota",
"credits",
"none"
],
"type": "string",
"description": "What this call drew on: the free daily allowance, account credits, or \"none\" (free calls)."
},
"auth_source": {
"enum": [
"bearer",
"api_key",
"anonymous"
],
"type": "string",
"description": "How you were recognised: \"bearer\" (signed-in connector or token), \"api_key\" (the argument) or \"anonymous\"."
},
"is_unlimited": {
"type": "boolean",
"description": "True for an internal account that is never charged."
},
"account_status": {
"type": [
"string",
"null"
],
"description": "\"created\" if this call created the account, otherwise \"matched\"; null without an account."
},
"credit_balance": {
"type": [
"integer",
"null"
],
"description": "Credits left on the account; null without an account."
},
"credits_charged": {
"type": "integer",
"description": "Credits this call spent."
},
"anonymous_quota_resets_at": {
"type": [
"string",
"null"
],
"description": "When the free allowance resets (next 00:00 UTC, ISO 8601)."
},
"anonymous_estimates_remaining_today": {
"type": [
"integer",
"null"
],
"description": "Free tier-1 estimates left today without an account; null for an account."
}
},
"description": "Who this call was billed to and what is left. Present on every Fermi result.",
"additionalProperties": true
},
"tier_id": {
"type": [
"integer",
"null"
],
"description": "The tier that produced it (1, 2 or 3)."
},
"provider": {
"type": [
"string",
"null"
],
"description": "AI provider that produced it."
},
"range_low": {
"type": [
"number",
"null"
],
"description": "Same as estimation_interval.low."
},
"created_at": {
"type": [
"string",
"null"
],
"description": "When it was made (ISO 8601)."
},
"disclaimer": {
"type": [
"string",
"null"
],
"description": "Not financial advice notice."
},
"next_steps": {
"type": [
"object",
"null"
],
"description": "Ready-made input for neumann, Kromatic's Monte Carlo simulation server. simulate.root_node (distribution_type and distribution_params) can be added to a neumann model as one uncertain input; when the family has no direct neumann match, simulate.solve_args are the arguments for neumann's convert_range_to_distribution instead. simulate.note says which, and simulate.mcp_url is the neumann server to call.",
"additionalProperties": true
},
"range_high": {
"type": [
"number",
"null"
],
"description": "Same as estimation_interval.high."
},
"request_id": {
"type": [
"string",
"null"
],
"description": "Id of this request, for support."
},
"assumptions": {
"type": [
"array",
"null"
],
"items": {
"type": "string"
},
"description": "The assumptions the estimate rests on."
},
"estimate_id": {
"type": [
"string",
"null"
],
"description": "Id of this estimate, for get_estimate_result and list_my_estimates."
},
"error_detail": {
"type": [
"string",
"null"
],
"description": "When failed: what went wrong."
},
"point_estimate": {
"type": [
"number",
"null"
],
"description": "The single best value. For a yes/no question asked with unit \"probability (0-1)\", this is the probability that the event happens (0.15 = 15%)."
},
"credits_charged": {
"type": [
"integer",
"null"
],
"description": "Credits this estimate cost (0 without an account)."
},
"confidence_level": {
"enum": [
"low",
"medium",
"high",
null
],
"type": [
"string",
"null"
],
"description": "Set only from coverage_probability: \"high\" for 0.95 or more, \"medium\" for 0.9 up to 0.95, \"low\" below 0.9. It is not a judgement of the estimate's quality."
},
"credits_refunded": {
"type": [
"integer",
"null"
],
"description": "When failed: credits given back."
},
"research_sources": {
"type": [
"array",
"null"
],
"items": {
"type": "object",
"additionalProperties": true
},
"description": "Tiers 2 and 3: web sources used (url, title, snippet, relevance_note)."
},
"reasoning_summary": {
"type": [
"string",
"null"
],
"description": "How the estimate was reached, in a few sentences."
},
"verification_flags": {
"type": [
"array",
"null"
],
"items": {
"type": "string"
},
"description": "Tier 3: issues the second check raised."
},
"distribution_family": {
"type": [
"string",
"null"
],
"description": "Shape of the fitted probability distribution: normal, log_normal (positive and right-skewed), uniform, constant, beta (values between 0 and 1, used for probabilities and rates), triangle, gamma, exponential or poisson (counts)."
},
"estimation_interval": {
"type": [
"object",
"null"
],
"properties": {
"low": {
"type": "number",
"description": "Lower end."
},
"high": {
"type": "number",
"description": "Upper end."
}
},
"description": "The range that should contain the true value with probability coverage_probability. For a probability question, both ends are probabilities between 0 and 1.",
"additionalProperties": true
},
"retry_after_seconds": {
"type": [
"integer",
"null"
],
"description": "When pending: seconds to wait before reading it again."
},
"coverage_probability": {
"type": [
"number",
"null"
],
"description": "Chance the interval contains the true value (0.9 = 90%)."
},
"verification_summary": {
"type": [
"string",
"null"
],
"description": "Tier 3: what a second check found."
},
"distribution_parameters": {
"type": [
"object",
"null"
],
"description": "Parameters of that distribution by name, e.g. alpha and beta for beta, mu and sigma for log_normal.",
"additionalProperties": {
"type": "number"
}
},
"decomposition_suggestions": {
"type": [
"array",
"null"
],
"items": {
"type": "object",
"additionalProperties": true
},
"description": "Sub-questions (sub_question, reasoning) worth estimating separately to sharpen this one. Each would be a new estimate call at the usual cost."
}
},
"description": "The estimate's status, plus its full result when completed. Fields that do not apply to the current status are null.",
"additionalProperties": true
},
"annotations": {
"title": "Get the result of an estimate you already requested",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}Lists the past estimates on your account, newest first. Use this when the user wants to find an earlier estimate ("what did I estimate last week?") or has lost an estimate_id. Returns estimates, each with estimate_id, created_at, question, unit, target_horizon, point_estimate, estimation_interval, distribution_family and status; and next_offset, the offset of the next page, or null on the last page. For an estimate's full detail, pass its estimate_id to get_estimate_result. Estimates made without an account cannot be listed. When the user is not signed in, call the tool as normal. If the user's app supports sign-in, it will ask the user to sign in, and you can call the tool again after that. Do not send the user to their connector or app settings to sign in first. Only if the user's app cannot sign in (a REST caller, or an agent with no connector), create an account with POST https://fermi.kromatic.com/api/v1/accounts (it returns an api_key once; a new account starts with 3 credits) and pass that key as the api_key argument. Cost: free, but account only: without an account the call is refused with a sign-in requirement.
| Parameter | Type | Description |
|---|---|---|
| limit | integer | How many estimates to return, 1 to 100. Default 20. |
| offset | integer | How many of the newest estimates to skip, for paging: pass the previous page's next_offset. Default 0. |
| api_key | any | Your Fermi API key. Required unless your connector is signed in. Get one with POST https://fermi.kromatic.com/api/v1/accounts. |
{
"name": "list_my_estimates",
"title": "List your past estimates (account only)",
"description": "Lists the past estimates on your account, newest first. Use this when the user wants to find an earlier estimate (\"what did I estimate last week?\") or has lost an estimate_id.\n\nReturns estimates, each with estimate_id, created_at, question, unit, target_horizon, point_estimate, estimation_interval, distribution_family and status; and next_offset, the offset of the next page, or null on the last page. For an estimate's full detail, pass its estimate_id to get_estimate_result.\n\nEstimates made without an account cannot be listed. When the user is not signed in, call the tool as normal. If the user's app supports sign-in, it will ask the user to sign in, and you can call the tool again after that. Do not send the user to their connector or app settings to sign in first. Only if the user's app cannot sign in (a REST caller, or an agent with no connector), create an account with POST https://fermi.kromatic.com/api/v1/accounts (it returns an api_key once; a new account starts with 3 credits) and pass that key as the api_key argument.\n\nCost: free, but account only: without an account the call is refused with a sign-in requirement.",
"inputSchema": {
"type": "object",
"title": "list_my_estimatesArguments",
"properties": {
"limit": {
"type": "integer",
"title": "Limit",
"default": 20,
"description": "How many estimates to return, 1 to 100. Default 20."
},
"offset": {
"type": "integer",
"title": "Offset",
"default": 0,
"description": "How many of the newest estimates to skip, for paging: pass the previous page's next_offset. Default 0."
},
"api_key": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Api Key",
"default": null,
"description": "Your Fermi API key. Required unless your connector is signed in. Get one with POST https://fermi.kromatic.com/api/v1/accounts."
}
}
},
"outputSchema": {
"type": "object",
"title": "ListMyEstimatesResult",
"properties": {
"billing": {
"type": "object",
"properties": {
"principal": {
"enum": [
"anonymous",
"account"
],
"type": "string",
"description": "\"anonymous\" (no account) or \"account\"."
},
"account_id": {
"type": [
"string",
"null"
],
"description": "Account id; null without an account."
},
"charged_to": {
"enum": [
"anonymous_quota",
"credits",
"none"
],
"type": "string",
"description": "What this call drew on: the free daily allowance, account credits, or \"none\" (free calls)."
},
"auth_source": {
"enum": [
"bearer",
"api_key",
"anonymous"
],
"type": "string",
"description": "How you were recognised: \"bearer\" (signed-in connector or token), \"api_key\" (the argument) or \"anonymous\"."
},
"is_unlimited": {
"type": "boolean",
"description": "True for an internal account that is never charged."
},
"account_status": {
"type": [
"string",
"null"
],
"description": "\"created\" if this call created the account, otherwise \"matched\"; null without an account."
},
"credit_balance": {
"type": [
"integer",
"null"
],
"description": "Credits left on the account; null without an account."
},
"credits_charged": {
"type": "integer",
"description": "Credits this call spent."
},
"anonymous_quota_resets_at": {
"type": [
"string",
"null"
],
"description": "When the free allowance resets (next 00:00 UTC, ISO 8601)."
},
"anonymous_estimates_remaining_today": {
"type": [
"integer",
"null"
],
"description": "Free tier-1 estimates left today without an account; null for an account."
}
},
"description": "Who this call was billed to and what is left. Present on every Fermi result.",
"additionalProperties": true
},
"estimates": {
"type": "array",
"items": {
"type": "object",
"properties": {
"unit": {
"type": "string"
},
"status": {
"enum": [
"pending",
"completed",
"failed"
],
"type": "string"
},
"question": {
"type": "string"
},
"created_at": {
"type": "string"
},
"estimate_id": {
"type": "string"
},
"point_estimate": {
"type": "number"
},
"target_horizon": {
"type": "string"
},
"distribution_family": {
"type": "string"
},
"estimation_interval": {
"type": "object",
"properties": {
"low": {
"type": "number",
"description": "Lower end."
},
"high": {
"type": "number",
"description": "Upper end."
}
},
"description": "The range that should contain the true value with probability coverage_probability. For a probability question, both ends are probabilities between 0 and 1.",
"additionalProperties": true
}
},
"additionalProperties": true
},
"description": "Your estimates, newest first."
},
"next_offset": {
"type": [
"integer",
"null"
],
"description": "Pass as offset for the next page; null on the last page."
}
},
"additionalProperties": true
},
"annotations": {
"title": "List your past estimates (account only)",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Records what actually happened for an estimate you made earlier, so the estimate can be compared with reality later (the calibration loop). Use this when the real-world outcome of a question you estimated is now known. Returns estimate_id and the stored actual (value, unit, recorded_at, notes, source_inputs). Calling it again with the same values stores the same record. Account only: you can record an actual only on an estimate made by your own account, and an anonymous call is refused with a sign-in requirement. To find an estimate_id, use list_my_estimates. When the user is not signed in, call the tool as normal. If the user's app supports sign-in, it will ask the user to sign in, and you can call the tool again after that. Do not send the user to their connector or app settings to sign in first. Only if the user's app cannot sign in (a REST caller, or an agent with no connector), create an account with POST https://fermi.kromatic.com/api/v1/accounts (it returns an api_key once; a new account starts with 3 credits) and pass that key as the api_key argument. Cost: free, but account only: recording never spends credits or a free estimate, and without an account the call is refused with a sign-in requirement.
| Parameter | Type | Description |
|---|---|---|
| api_key | any | Your Fermi API key. Required unless your connector is signed in. Get one with POST https://fermi.kromatic.com/api/v1/accounts. |
| actual_unit | any | Unit of actual_value, if different from the estimate's own unit. Defaults to the estimate's unit. |
| estimate_id * | string | The estimate_id returned by estimate_quantity_or_probability or listed by list_my_estimates. |
| actual_notes | any | Optional free-text notes about the actual outcome. |
| actual_value * |
Sends a short report about Fermi to the team that builds it. Use this when the user wants to report a problem, an error, an estimate that looks wrong, something no Fermi tool can do, a missing option, or copy that was confusing, or when the user says yes after you offered to send a report. The team reads every report and uses it to fix problems and decide what to build. Send a report only when the user asks for one or agrees to one. When the user asks you to tell the Fermi team about something, that request is their go-ahead: send the report without asking again. If a report might help, ask first, for example: "Want me to send Fermi's team a short note about this?" If the user says no or does not answer, do not send it. Put several points in one report. Only message is required. Fill in the other fields that apply so the team can act without guessing. Write in your own words. Never include personal data (names, email addresses, company names), the user's own numbers or private context, secrets or keys, or the conversation transcript; describe the question and any wrong result in general terms instead. Add sender_email only if the user wants a reply; no copy is sent to it. If you are signed in, your account id is included. Returns status "sent". Cost: free. Works without an account.
| Parameter | Type | Description |
|---|---|---|
| impact | any | How much it cost the user. blocked: could not finish. workaround: finished another way. minor: an annoyance. none: no cost, for example praise. |
| api_key | any | Optional Fermi API key. If given, the account id is attached so the team can look into it. |
| message * | string | The report, in your own words, 10 to 5,000 characters. Say what you tried, what happened and what you expected. Leave out names, companies, the user's numbers or private context, and the conversation itself: write "a market-size question", not the question's details. |
| expected | any |
Charges the user's saved payment card to buy Fermi credits, or, if no card is saved yet, returns a checkout link instead and charges nothing. Use this when the user asks to buy or top up Fermi credits, including before the dollar amount is named and while the user is not signed in. Never call it on your own initiative. The user's authorization: before the call, tell the user the number of credits and the dollar charge (credits times $0.50; for example 10 credits is $5.00). The user's yes to that, or a request that already names both the credits and the dollar amount, authorizes that one call: make it then, without asking again, and make it again after a sign-in. It does not authorize a second purchase. A request that names only the credits ("buy 10 credits") is still a request for this tool: state the dollar charge, and make the call on the user's yes. A call refused because the user is not signed in charges nothing, so the call made again after the sign-in is still that one purchase. Returns credits_purchased, new_balance and payment_status after a charge. With no saved card, the first call charges nothing: it returns a checkout_url where the user saves a card in a browser, so that call needs no separate warning about a charge. Then call this again. One call can buy at most 100 credits; a larger request is refused and nothing is charged. Credits pay for estimates on an account (see estimate_quantity_or_probability for tier costs). To see the balance first, use check_account_and_credits. When the user is not signed in, call the tool as normal. If the user's app supports sign-in, it will ask the user to sign in, and you can call the tool again after that. Do not send the user to their connector or app settings to sign in first. Only if the user's app cannot sign in (a REST caller, or an agent with no connector), create an account with POST https://fermi.kromatic.com/api/v1/accounts (it returns an api_key once; a new account starts with 3 credits) and pass that key as the api_key argument. Cost: real money -- charges the saved card $0.50 per credit. Needs an account.
| Parameter | Type | Description |
|---|---|---|
| api_key | any | Your Fermi API key. Omit it if your connector is signed in with permission to buy credits. |
| credits |
Reports who you are to Fermi and what you can spend: whether you are using Fermi without an account or with one, your credit balance or free estimates left today, and what each tier costs. Use this before a paid estimate to check the price or balance, after one to see what is left, or when the user asks about credits, free estimates or pricing. Returns billing (principal "anonymous" or "account", how you authenticated, credit_balance, anonymous_estimates_remaining_today and anonymous_quota_resets_at), tiers (credit_cost, delivery "sync" or "async" and whether it is available without an account, per tier), price_per_credit_usd, and recent_transactions (the account's last 10 credit movements; null without an account). Cost: free. Checking never spends credits and never uses up a free estimate.
| Parameter | Type | Description |
|---|---|---|
| api_key | any | Your Fermi API key. Omit it to ask about your connector's signed-in account, or about your free allowance without an account. |
{
"name": "check_account_and_credits",
"title": "Check your account, credits and free estimates left",
"description": "Reports who you are to Fermi and what you can spend: whether you are using Fermi without an account or with one, your credit balance or free estimates left today, and what each tier costs. Use this before a paid estimate to check the price or balance, after one to see what is left, or when the user asks about credits, free estimates or pricing.\n\nReturns billing (principal \"anonymous\" or \"account\", how you authenticated, credit_balance, anonymous_estimates_remaining_today and anonymous_quota_resets_at), tiers (credit_cost, delivery \"sync\" or \"async\" and whether it is available without an account, per tier), price_per_credit_usd, and recent_transactions (the account's last 10 credit movements; null without an account).\n\nCost: free. Checking never spends credits and never uses up a free estimate.",
"inputSchema": {
"type": "object",
"title": "check_account_and_creditsArguments",
"properties": {
"api_key": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Api Key",
"default": null,
"description": "Your Fermi API key. Omit it to ask about your connector's signed-in account, or about your free allowance without an account."
}
}
},
"outputSchema": {
"type": "object",
"title": "CheckAccountAndCreditsResult",
"properties": {
"tiers": {
"type": "array",
"items": {
"type": "object",
"properties": {
"tier_id": {
"type": "integer"
},
"delivery": {
"type": "string",
"description": "\"sync\" (answer in the same call) or \"async\" (collect with get_estimate_result)."
},
"credit_cost": {
"type": "integer",
"description": "Credits one estimate at this tier costs."
},
"supports_anonymous_trial": {
"type": "boolean",
"description": "True if usable without an account."
}
},
"additionalProperties": true
},
"description": "Each tier's price and availability."
},
"billing": {
"type": "object",
"properties": {
"principal": {
"enum": [
"anonymous",
"account"
],
"type": "string",
"description": "\"anonymous\" (no account) or \"account\"."
},
"account_id": {
"type": [
"string",
"null"
],
"description": "Account id; null without an account."
},
"charged_to": {
"enum": [
"anonymous_quota",
"credits",
"none"
],
"type": "string",
"description": "What this call drew on: the free daily allowance, account credits, or \"none\" (free calls)."
},
"auth_source": {
"enum": [
"bearer",
"api_key",
"anonymous"
],
"type": "string",
"description": "How you were recognised: \"bearer\" (signed-in connector or token), \"api_key\" (the argument) or \"anonymous\"."
},
"is_unlimited": {
"type": "boolean",
"description": "True for an internal account that is never charged."
},
"account_status": {
"type": [
"string",
"null"
],
"description": "\"created\" if this call created the account, otherwise \"matched\"; null without an account."
},
"credit_balance": {
"type": [
"integer",
"null"
],
"description": "Credits left on the account; null without an account."
},
"credits_charged": {
"type": "integer",
"description": "Credits this call spent."
},
"anonymous_quota_resets_at": {
"type": [
"string",
"null"
],
"description": "When the free allowance resets (next 00:00 UTC, ISO 8601)."
},
"anonymous_estimates_remaining_today": {
"type": [
"integer",
"null"
],
"description": "Free tier-1 estimates left today without an account; null for an account."
}
},
"description": "Who this call was billed to and what is left. Present on every Fermi result.",
"additionalProperties": true
},
"recent_transactions": {
"type": [
"array",
"null"
],
"items": {
"type": "object",
"additionalProperties": true
},
"description": "The account's last 10 credit movements (entry_type, amount, request_id, created_at); null without an account."
},
"price_per_credit_usd": {
"type": "number",
"description": "What one credit costs to buy, in US dollars."
}
},
"additionalProperties": true
},
"annotations": {
"title": "Check your account, credits and free estimates left",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Fermi answers forecasting questions with calibrated estimates. Call estimate_quantity_or_probability for either kind of question: how much / how many ("How many hotel rooms will Lisbon have in 2030?"), or will an event happen by a date ("Will any EU country make a four-day work week law before 2030?" -- set unit to "probability (0-1)"; point_estimate is the probability). Each answer is a range plus a probability distribution, with reasoning and assumptions. Without an account you get 3 free tier-1 estimates per UTC day. With an account, tier 1/2/3 costs 1/3/10 credits at $0.50 per credit; tier 3 runs in the background, so collect it with get_estimate_result. When the user wants an account-only call and is not signed in, call the tool as normal. If the user's app supports sign-in, it will ask the user to sign in, and you can call the tool again after that. Do not send the user to their connector or app settings to sign in first. Only if the user's app cannot sign in (a REST caller, or an agent with no connector), create an account with POST https://fermi.kromatic.com/api/v1/accounts (it returns an api_key once; a new account starts with 3 credits) and pass that key as the api_key argument. list_my_estimates finds earlier estimates (account only). record_actual records what actually happened for one of your estimates (account only). check_account_and_credits shows your balance, free estimates left and prices, for free. buy_credits charges the user's saved card, at most 100 credits per call: state the credits and the dollar charge first, and the user's yes (or a request that already names both) authorizes that one call. A request that names only the credits still goes to buy_credits: state the charge, then call it on the yes. send_feedback emails the Fermi team. To simulate a model built from several estimates, pass each estimate's next_steps.simulate to neumann, Kromatic's Monte Carlo simulation server. Feedback helps the Fermi team, and it is sent only when the user asks for it or agrees to it. At a natural moment (a Fermi tool errored, a result looks wrong, the user needs something Fermi cannot do, or the work is wrapping up), offer to send a report as a question, for example: "Want me to send Fermi's team a short note about this?" Call send_feedback only after the user says yes, with a short report in your own words (no personal data, secrets or transcript).
Names, descriptions and schemas were written by the server and are shown as data.
Pattern matches on what the server declares. They show what it says it can do, not what it does.
None yet. Changes appear after a server's second daily check.
| target_horizon * | string | When the answer applies. Prefer an absolute date or year: "by 2027-06-30", "2030", "end of 2028", or "current" for a value today. Relative phrases like "in the next 12 months" are read from the server's current UTC date. |
| requested_tier_id | integer | 1 = fast: answers in the same call, no web search; the only tier without an account. 2 = grounded: answers in the same call, searches the web; account only. 3 = deep research: runs in the background and searches the web; returns status "pending" and an estimate_id to read later with get_estimate_res… |
| coverage_probability | number | How likely the returned interval is to contain the true value, strictly between 0 and 1. Default 0.9, a 90% interval. Use 0.95 for a "95% interval", 0.8 for an "80% range". |
| disclaimer_acknowledged * | boolean | Must be true. Confirms the result is a probabilistic estimate for decision support, not financial, medical or legal advice. A call with false is refused. |
| number |
| The actual outcome, in the same unit as the estimate. |
| actual_recorded_at | any | When the actual outcome occurred or was observed, as an ISO 8601 date-time. Defaults to now. |
{
"name": "record_actual",
"title": "Record the actual outcome (account only)",
"description": "Records what actually happened for an estimate you made earlier, so the estimate can be compared with reality later (the calibration loop). Use this when the real-world outcome of a question you estimated is now known.\n\nReturns estimate_id and the stored actual (value, unit, recorded_at, notes, source_inputs). Calling it again with the same values stores the same record.\n\nAccount only: you can record an actual only on an estimate made by your own account, and an anonymous call is refused with a sign-in requirement. To find an estimate_id, use list_my_estimates. When the user is not signed in, call the tool as normal. If the user's app supports sign-in, it will ask the user to sign in, and you can call the tool again after that. Do not send the user to their connector or app settings to sign in first. Only if the user's app cannot sign in (a REST caller, or an agent with no connector), create an account with POST https://fermi.kromatic.com/api/v1/accounts (it returns an api_key once; a new account starts with 3 credits) and pass that key as the api_key argument.\n\nCost: free, but account only: recording never spends credits or a free estimate, and without an account the call is refused with a sign-in requirement.",
"inputSchema": {
"type": "object",
"title": "record_actualArguments",
"required": [
"estimate_id",
"actual_value"
],
"properties": {
"api_key": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Api Key",
"default": null,
"description": "Your Fermi API key. Required unless your connector is signed in. Get one with POST https://fermi.kromatic.com/api/v1/accounts."
},
"actual_unit": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Actual Unit",
"default": null,
"description": "Unit of actual_value, if different from the estimate's own unit. Defaults to the estimate's unit."
},
"estimate_id": {
"type": "string",
"title": "Estimate Id",
"description": "The estimate_id returned by estimate_quantity_or_probability or listed by list_my_estimates."
},
"actual_notes": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Actual Notes",
"default": null,
"description": "Optional free-text notes about the actual outcome."
},
"actual_value": {
"type": "number",
"title": "Actual Value",
"description": "The actual outcome, in the same unit as the estimate."
},
"actual_recorded_at": {
"anyOf": [
{
"type": "string",
"format": "date-time"
},
{
"type": "null"
}
],
"title": "Actual Recorded At",
"default": null,
"description": "When the actual outcome occurred or was observed, as an ISO 8601 date-time. Defaults to now."
}
}
},
"outputSchema": {
"type": "object",
"title": "RecordActualResult",
"properties": {
"actual": {
"type": "object",
"properties": {
"unit": {
"type": "string"
},
"notes": {
"type": [
"string",
"null"
]
},
"value": {
"type": "number"
},
"recorded_at": {
"type": "string"
},
"source_inputs": {
"type": "array",
"items": {
"type": "object"
}
}
},
"description": "The stored actual outcome.",
"additionalProperties": true
},
"billing": {
"type": "object",
"properties": {
"principal": {
"enum": [
"anonymous",
"account"
],
"type": "string",
"description": "\"anonymous\" (no account) or \"account\"."
},
"account_id": {
"type": [
"string",
"null"
],
"description": "Account id; null without an account."
},
"charged_to": {
"enum": [
"anonymous_quota",
"credits",
"none"
],
"type": "string",
"description": "What this call drew on: the free daily allowance, account credits, or \"none\" (free calls)."
},
"auth_source": {
"enum": [
"bearer",
"api_key",
"anonymous"
],
"type": "string",
"description": "How you were recognised: \"bearer\" (signed-in connector or token), \"api_key\" (the argument) or \"anonymous\"."
},
"is_unlimited": {
"type": "boolean",
"description": "True for an internal account that is never charged."
},
"account_status": {
"type": [
"string",
"null"
],
"description": "\"created\" if this call created the account, otherwise \"matched\"; null without an account."
},
"credit_balance": {
"type": [
"integer",
"null"
],
"description": "Credits left on the account; null without an account."
},
"credits_charged": {
"type": "integer",
"description": "Credits this call spent."
},
"anonymous_quota_resets_at": {
"type": [
"string",
"null"
],
"description": "When the free allowance resets (next 00:00 UTC, ISO 8601)."
},
"anonymous_estimates_remaining_today": {
"type": [
"integer",
"null"
],
"description": "Free tier-1 estimates left today without an account; null for an account."
}
},
"description": "Who this call was billed to and what is left. Present on every Fermi result.",
"additionalProperties": true
},
"estimate_id": {
"type": "string"
}
},
"additionalProperties": true
},
"annotations": {
"title": "Record the actual outcome (account only)",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}| What you expected to happen instead. |
| tool_name | any | The Fermi tool the report is about, for example estimate_quantity_or_probability. |
| user_goal | any | What the user was trying to do, as a general task in a few words, for example "forecast a market size". No names, company names or figures. |
| error_text | any | The error code and message the tool returned, if any. Remove any personal data or private context it quotes. |
| suggestion | any | What would have helped: a fix, a new feature or clearer wording. |
| estimate_id | any | The estimate_id the report is about, if an estimate was involved. |
| sender_name | any | The user's name. Optional; only with sender_email. |
| sender_email | any | Optional. The user's email address, only if they want a reply and gave it to you. Do not invent one. No copy is sent to it. |
| feedback_type | string | What kind of report this is. bug: a tool errored or broke. wrong_result: it ran but the estimate looks wrong. unmet_use_case: the user needed something no tool does. feature_request: a missing option or parameter. confusing: a description, output or error was hard to understand. praise: something wo… |
| what_happened | any | What actually happened, in general terms, without the user's numbers. |
{
"name": "send_feedback",
"title": "Send feedback or a bug report to the Fermi team",
"description": "Sends a short report about Fermi to the team that builds it. Use this when the user wants to report a problem, an error, an estimate that looks wrong, something no Fermi tool can do, a missing option, or copy that was confusing, or when the user says yes after you offered to send a report. The team reads every report and uses it to fix problems and decide what to build.\n\nSend a report only when the user asks for one or agrees to one. When the user asks you to tell the Fermi team about something, that request is their go-ahead: send the report without asking again. If a report might help, ask first, for example: \"Want me to send Fermi's team a short note about this?\" If the user says no or does not answer, do not send it.\n\nPut several points in one report. Only message is required. Fill in the other fields that apply so the team can act without guessing. Write in your own words. Never include personal data (names, email addresses, company names), the user's own numbers or private context, secrets or keys, or the conversation transcript; describe the question and any wrong result in general terms instead. Add sender_email only if the user wants a reply; no copy is sent to it. If you are signed in, your account id is included. Returns status \"sent\".\n\nCost: free. Works without an account.",
"inputSchema": {
"type": "object",
"title": "send_feedbackArguments",
"required": [
"message"
],
"properties": {
"impact": {
"anyOf": [
{
"enum": [
"blocked",
"workaround",
"minor",
"none"
],
"type": "string"
},
{
"type": "null"
}
],
"title": "Impact",
"default": null,
"description": "How much it cost the user. blocked: could not finish. workaround: finished another way. minor: an annoyance. none: no cost, for example praise."
},
"api_key": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Api Key",
"default": null,
"description": "Optional Fermi API key. If given, the account id is attached so the team can look into it."
},
"message": {
"type": "string",
"title": "Message",
"description": "The report, in your own words, 10 to 5,000 characters. Say what you tried, what happened and what you expected. Leave out names, companies, the user's numbers or private context, and the conversation itself: write \"a market-size question\", not the question's details."
},
"expected": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Expected",
"default": null,
"description": "What you expected to happen instead."
},
"tool_name": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Tool Name",
"default": null,
"description": "The Fermi tool the report is about, for example estimate_quantity_or_probability."
},
"user_goal": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "User Goal",
"default": null,
"description": "What the user was trying to do, as a general task in a few words, for example \"forecast a market size\". No names, company names or figures."
},
"error_text": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Error Text",
"default": null,
"description": "The error code and message the tool returned, if any. Remove any personal data or private context it quotes."
},
"suggestion": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Suggestion",
"default": null,
"description": "What would have helped: a fix, a new feature or clearer wording."
},
"estimate_id": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Estimate Id",
"default": null,
"description": "The estimate_id the report is about, if an estimate was involved."
},
"sender_name": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Sender Name",
"default": null,
"description": "The user's name. Optional; only with sender_email."
},
"sender_email": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Sender Email",
"default": null,
"description": "Optional. The user's email address, only if they want a reply and gave it to you. Do not invent one. No copy is sent to it."
},
"feedback_type": {
"enum": [
"bug",
"unmet_use_case",
"feature_request",
"confusing",
"wrong_result",
"praise",
"other"
],
"type": "string",
"title": "Feedback Type",
"default": "other",
"description": "What kind of report this is. bug: a tool errored or broke. wrong_result: it ran but the estimate looks wrong. unmet_use_case: the user needed something no tool does. feature_request: a missing option or parameter. confusing: a description, output or error was hard to understand. praise: something worked well. other: anything else (the default)."
},
"what_happened": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "What Happened",
"default": null,
"description": "What actually happened, in general terms, without the user's numbers."
}
}
},
"outputSchema": {
"type": "object",
"title": "SendFeedbackResult",
"properties": {
"status": {
"type": "string",
"description": "\"sent\"."
},
"billing": {
"type": "object",
"properties": {
"principal": {
"enum": [
"anonymous",
"account"
],
"type": "string",
"description": "\"anonymous\" (no account) or \"account\"."
},
"account_id": {
"type": [
"string",
"null"
],
"description": "Account id; null without an account."
},
"charged_to": {
"enum": [
"anonymous_quota",
"credits",
"none"
],
"type": "string",
"description": "What this call drew on: the free daily allowance, account credits, or \"none\" (free calls)."
},
"auth_source": {
"enum": [
"bearer",
"api_key",
"anonymous"
],
"type": "string",
"description": "How you were recognised: \"bearer\" (signed-in connector or token), \"api_key\" (the argument) or \"anonymous\"."
},
"is_unlimited": {
"type": "boolean",
"description": "True for an internal account that is never charged."
},
"account_status": {
"type": [
"string",
"null"
],
"description": "\"created\" if this call created the account, otherwise \"matched\"; null without an account."
},
"credit_balance": {
"type": [
"integer",
"null"
],
"description": "Credits left on the account; null without an account."
},
"credits_charged": {
"type": "integer",
"description": "Credits this call spent."
},
"anonymous_quota_resets_at": {
"type": [
"string",
"null"
],
"description": "When the free allowance resets (next 00:00 UTC, ISO 8601)."
},
"anonymous_estimates_remaining_today": {
"type": [
"integer",
"null"
],
"description": "Free tier-1 estimates left today without an account; null for an account."
}
},
"description": "Who this call was billed to and what is left. Present on every Fermi result.",
"additionalProperties": true
},
"message": {
"type": "string",
"description": "Confirmation text, including whether the team can reply."
}
},
"additionalProperties": true
},
"annotations": {
"title": "Send feedback or a bug report to the Fermi team",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}| integer |
| How many credits to buy, 1 to 100. Each is charged to the saved card at the price in this tool's cost line. |
{
"name": "buy_credits",
"title": "Buy Fermi credits with your saved card",
"description": "Charges the user's saved payment card to buy Fermi credits, or, if no card is saved yet, returns a checkout link instead and charges nothing. Use this when the user asks to buy or top up Fermi credits, including before the dollar amount is named and while the user is not signed in. Never call it on your own initiative.\n\nThe user's authorization: before the call, tell the user the number of credits and the dollar charge (credits times $0.50; for example 10 credits is $5.00). The user's yes to that, or a request that already names both the credits and the dollar amount, authorizes that one call: make it then, without asking again, and make it again after a sign-in. It does not authorize a second purchase. A request that names only the credits (\"buy 10 credits\") is still a request for this tool: state the dollar charge, and make the call on the user's yes. A call refused because the user is not signed in charges nothing, so the call made again after the sign-in is still that one purchase.\n\nReturns credits_purchased, new_balance and payment_status after a charge. With no saved card, the first call charges nothing: it returns a checkout_url where the user saves a card in a browser, so that call needs no separate warning about a charge. Then call this again.\n\nOne call can buy at most 100 credits; a larger request is refused and nothing is charged.\n\nCredits pay for estimates on an account (see estimate_quantity_or_probability for tier costs). To see the balance first, use check_account_and_credits. When the user is not signed in, call the tool as normal. If the user's app supports sign-in, it will ask the user to sign in, and you can call the tool again after that. Do not send the user to their connector or app settings to sign in first. Only if the user's app cannot sign in (a REST caller, or an agent with no connector), create an account with POST https://fermi.kromatic.com/api/v1/accounts (it returns an api_key once; a new account starts with 3 credits) and pass that key as the api_key argument.\n\nCost: real money -- charges the saved card $0.50 per credit. Needs an account.",
"inputSchema": {
"type": "object",
"title": "buy_creditsArguments",
"required": [
"credits"
],
"properties": {
"api_key": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Api Key",
"default": null,
"description": "Your Fermi API key. Omit it if your connector is signed in with permission to buy credits."
},
"credits": {
"type": "integer",
"title": "Credits",
"maximum": 100,
"description": "How many credits to buy, 1 to 100. Each is charged to the saved card at the price in this tool's cost line."
}
}
},
"outputSchema": {
"type": "object",
"title": "BuyCreditsResult",
"properties": {
"billing": {
"type": "object",
"properties": {
"principal": {
"enum": [
"anonymous",
"account"
],
"type": "string",
"description": "\"anonymous\" (no account) or \"account\"."
},
"account_id": {
"type": [
"string",
"null"
],
"description": "Account id; null without an account."
},
"charged_to": {
"enum": [
"anonymous_quota",
"credits",
"none"
],
"type": "string",
"description": "What this call drew on: the free daily allowance, account credits, or \"none\" (free calls)."
},
"auth_source": {
"enum": [
"bearer",
"api_key",
"anonymous"
],
"type": "string",
"description": "How you were recognised: \"bearer\" (signed-in connector or token), \"api_key\" (the argument) or \"anonymous\"."
},
"is_unlimited": {
"type": "boolean",
"description": "True for an internal account that is never charged."
},
"account_status": {
"type": [
"string",
"null"
],
"description": "\"created\" if this call created the account, otherwise \"matched\"; null without an account."
},
"credit_balance": {
"type": [
"integer",
"null"
],
"description": "Credits left on the account; null without an account."
},
"credits_charged": {
"type": "integer",
"description": "Credits this call spent."
},
"anonymous_quota_resets_at": {
"type": [
"string",
"null"
],
"description": "When the free allowance resets (next 00:00 UTC, ISO 8601)."
},
"anonymous_estimates_remaining_today": {
"type": [
"integer",
"null"
],
"description": "Free tier-1 estimates left today without an account; null for an account."
}
},
"description": "Who this call was billed to and what is left. Present on every Fermi result.",
"additionalProperties": true
},
"new_balance": {
"type": "integer",
"description": "Credit balance after it."
},
"checkout_url": {
"type": "string",
"description": "No card saved: open this in a browser to save one, then call buy_credits again."
},
"credits_added": {
"type": "integer",
"description": "0 when a checkout_url is returned instead."
},
"payment_status": {
"type": "string",
"description": "Card payment status."
},
"credits_purchased": {
"type": "integer",
"description": "Credits bought by this call."
}
},
"description": "Either a completed purchase (credits_purchased, new_balance, payment_status) or, with no saved card, a checkout_url and credits_added 0.",
"additionalProperties": true
},
"annotations": {
"title": "Buy Fermi credits with your saved card",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": true
}
}