SavingsResponse
What agents avoided in one repository over one window. Every figure comes from the canonical savings ledger through the one core report service, so this response, the repository overview headline and ``repowise saved`` cannot disagree -- which the three of them did, in four separate ways, when each aggregated and priced the ledger for itself. Three things it deliberately keeps apart. *Measured* reductions are known before/after sizes from an operation that really ran; *inferred* avoidance is a documented counterfactual. *Priced* and *unpriced* tokens are split because an event records the rate it was worth at the time and some events carry none, and reporting a total as though it were all priced would be a guess. *Observed opportunities* are things that could have been saved and were not: they never enter the headline. ``available`` is False when the repository has no savings sidecar at all, which means "nothing has been measured here" and is different from a measured zero.
Properties
| Name | Type | Description |
|---|---|---|
| available | boolean | |
| window_days | object | |
| as_of | string | |
| first_event_at | object | |
| last_event_at | object | |
| unique_events | integer | |
| successful_or_usable_partial_events | integer | |
| saving_interactions | integer | |
| mcp_queries_answered | integer | |
| dead_ends | integer | |
| saved_input_tokens | integer | |
| measured_saved_input_tokens | integer | |
| inferred_saved_input_tokens | integer | |
| priced_saved_input_tokens | integer | |
| unpriced_saved_input_tokens | integer | |
| priced_input_savings_usd | number | |
| saved_output_tokens | object | |
| priced_saved_output_tokens | integer | |
| unpriced_saved_output_tokens | integer | |
| priced_output_savings_usd | number | |
| baseline_events | integer | |
| reducing_events | integer | |
| baseline_input_tokens | integer | |
| baseline_saved_input_tokens | integer | |
| input_reduction_ratio | object | |
| input_reduction_ratio_p90 | object | |
| per_operation | array | |
| per_surface | array | |
| per_agent | array | |
| per_model | array | |
| per_day | array | |
| opportunity_count | integer | |
| opportunity_tokens_excluded | integer | |
| per_opportunity_kind | array | |
| missed_events | integer | |
| missed_tokens_est | integer | |
| missed_window_days | number | |
| reread_events | integer | |
| reread_tokens_est | integer |
JSON Schema
{
"$schema": "https://json-schema.org/draft/2020-12/schema",
"$id": "https://artifacts.apis.io/repowise-dev/json-schema/repowise-dev-savings-response-schema.json",
"title": "SavingsResponse",
"description": "What agents avoided in one repository over one window.\n\nEvery figure comes from the canonical savings ledger through the one core\nreport service, so this response, the repository overview headline and\n``repowise saved`` cannot disagree -- which the three of them did, in four\nseparate ways, when each aggregated and priced the ledger for itself.\n\nThree things it deliberately keeps apart. *Measured* reductions are known\nbefore/after sizes from an operation that really ran; *inferred* avoidance\nis a documented counterfactual. *Priced* and *unpriced* tokens are split\nbecause an event records the rate it was worth at the time and some events\ncarry none, and reporting a total as though it were all priced would be a\nguess. *Observed opportunities* are things that could have been saved and\nwere not: they never enter the headline.\n\n``available`` is False when the repository has no savings sidecar at all,\nwhich means \"nothing has been measured here\" and is different from a\nmeasured zero.",
"x-generated": "2026-10-30",
"x-method": "derived",
"x-generator": "derive-json-schema.py",
"x-source": "openapi/repowise-dev-openapi.json#/components/schemas/SavingsResponse",
"properties": {
"available": {
"type": "boolean",
"title": "Available"
},
"window_days": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"title": "Window Days"
},
"as_of": {
"type": "string",
"title": "As Of",
"default": ""
},
"first_event_at": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "First Event At"
},
"last_event_at": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Last Event At"
},
"unique_events": {
"type": "integer",
"title": "Unique Events",
"default": 0
},
"successful_or_usable_partial_events": {
"type": "integer",
"title": "Successful Or Usable Partial Events",
"default": 0
},
"saving_interactions": {
"type": "integer",
"title": "Saving Interactions",
"default": 0
},
"mcp_queries_answered": {
"type": "integer",
"title": "Mcp Queries Answered",
"default": 0
},
"dead_ends": {
"type": "integer",
"title": "Dead Ends",
"default": 0
},
"saved_input_tokens": {
"type": "integer",
"title": "Saved Input Tokens",
"default": 0
},
"measured_saved_input_tokens": {
"type": "integer",
"title": "Measured Saved Input Tokens",
"default": 0
},
"inferred_saved_input_tokens": {
"type": "integer",
"title": "Inferred Saved Input Tokens",
"default": 0
},
"priced_saved_input_tokens": {
"type": "integer",
"title": "Priced Saved Input Tokens",
"default": 0
},
"unpriced_saved_input_tokens": {
"type": "integer",
"title": "Unpriced Saved Input Tokens",
"default": 0
},
"priced_input_savings_usd": {
"type": "number",
"title": "Priced Input Savings Usd",
"default": 0.0
},
"saved_output_tokens": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"title": "Saved Output Tokens"
},
"priced_saved_output_tokens": {
"type": "integer",
"title": "Priced Saved Output Tokens",
"default": 0
},
"unpriced_saved_output_tokens": {
"type": "integer",
"title": "Unpriced Saved Output Tokens",
"default": 0
},
"priced_output_savings_usd": {
"type": "number",
"title": "Priced Output Savings Usd",
"default": 0.0
},
"baseline_events": {
"type": "integer",
"title": "Baseline Events",
"default": 0
},
"reducing_events": {
"type": "integer",
"title": "Reducing Events",
"default": 0
},
"baseline_input_tokens": {
"type": "integer",
"title": "Baseline Input Tokens",
"default": 0
},
"baseline_saved_input_tokens": {
"type": "integer",
"title": "Baseline Saved Input Tokens",
"default": 0
},
"input_reduction_ratio": {
"anyOf": [
{
"type": "number"
},
{
"type": "null"
}
],
"title": "Input Reduction Ratio"
},
"input_reduction_ratio_p90": {
"anyOf": [
{
"type": "number"
},
{
"type": "null"
}
],
"title": "Input Reduction Ratio P90"
},
"per_operation": {
"items": {
"$ref": "#/$defs/SavingsBreakdownRow"
},
"type": "array",
"title": "Per Operation",
"default": []
},
"per_surface": {
"items": {
"$ref": "#/$defs/SavingsBreakdownRow"
},
"type": "array",
"title": "Per Surface",
"default": []
},
"per_agent": {
"items": {
"$ref": "#/$defs/SavingsAgentRow"
},
"type": "array",
"title": "Per Agent",
"default": []
},
"per_model": {
"items": {
"$ref": "#/$defs/SavingsBreakdownRow"
},
"type": "array",
"title": "Per Model",
"default": []
},
"per_day": {
"items": {
"$ref": "#/$defs/SavingsBreakdownRow"
},
"type": "array",
"title": "Per Day",
"default": []
},
"opportunity_count": {
"type": "integer",
"title": "Opportunity Count",
"default": 0
},
"opportunity_tokens_excluded": {
"type": "integer",
"title": "Opportunity Tokens Excluded",
"default": 0
},
"per_opportunity_kind": {
"items": {
"$ref": "#/$defs/SavingsOpportunityRow"
},
"type": "array",
"title": "Per Opportunity Kind",
"default": []
},
"missed_events": {
"type": "integer",
"title": "Missed Events",
"default": 0
},
"missed_tokens_est": {
"type": "integer",
"title": "Missed Tokens Est",
"default": 0
},
"missed_window_days": {
"type": "number",
"title": "Missed Window Days",
"default": 0.0
},
"reread_events": {
"type": "integer",
"title": "Reread Events",
"default": 0
},
"reread_tokens_est": {
"type": "integer",
"title": "Reread Tokens Est",
"default": 0
}
},
"type": "object",
"required": [
"available"
],
"$defs": {
"SavingsAgentRow": {
"properties": {
"agent": {
"type": "string",
"title": "Agent"
},
"agent_display_name": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Agent Display Name"
},
"events": {
"type": "integer",
"title": "Events",
"default": 0
},
"saved_input_tokens": {
"type": "integer",
"title": "Saved Input Tokens",
"default": 0
}
},
"type": "object",
"required": [
"agent"
],
"title": "SavingsAgentRow",
"description": "One agent's savings, labelled from the identity registry."
},
"SavingsBreakdownRow": {
"properties": {
"group": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Group"
},
"events": {
"type": "integer",
"title": "Events",
"default": 0
},
"saved_input_tokens": {
"type": "integer",
"title": "Saved Input Tokens",
"default": 0
}
},
"type": "object",
"title": "SavingsBreakdownRow",
"description": "One bucket of a savings breakdown.\n\n``group`` is nullable because one bucket genuinely has no name: an event\nwritten before its rate could be resolved has no model, and that is a real\nquantity -- how much saving carries no pricing evidence -- rather than a\ngap to drop from the list."
},
"SavingsOpportunityRow": {
"properties": {
"kind": {
"type": "string",
"title": "Kind"
},
"observations": {
"type": "integer",
"title": "Observations",
"default": 0
},
"estimated_potential_input_tokens": {
"type": "integer",
"title": "Estimated Potential Input Tokens",
"default": 0
}
},
"type": "object",
"required": [
"kind"
],
"title": "SavingsOpportunityRow",
"description": "One kind of observed opportunity. Never part of achieved savings."
}
}
}
Work with this as data
Every JSON Schema here is available over the APIs.io API and to AI agents over MCP.
MCP server
One button, every client — Claude, Cursor, VS Code and the rest.
https://apis.io/mcp
Tools for schemas
4 MCP tools reach this
find_json_schemasBrowse and filter every JSON Schema in the catalog.apis_io_searchSTART HERE — APIs, providers and tags for one query, each with its total.resolveTurn a domain, URL or GitHub org into the provider it belongs to.find_cohortsEvery scored population of providers in the catalog.
Call it yourself
curl for this page
curl "https://apis.io/api/v1/json-schemas/repowise-dev-savings-response"
curl "https://apis.io/api/v1/json-schemas?limit=25"
Discovery needs no key. Ratings and market analysis are Pro.
Get an API key
Free tier, no form to fill in. Signing in shares your email address with us — we store it to create your key and to recognise you if you sign in with another provider. See our Privacy Policy and Terms.
A second provider on the same verified email joins the account you already have.