Workspace Intelligence / JSON Schema

Workspai Workspace Intelligence Evaluation Comparison

A task-aligned comparison between two provenance-aware Workspace Intelligence evaluation runs.

Schema version: workspace-intelligence-evaluation-comparison.v1
Contract path: contracts/workspace-intelligence/workspace-intelligence-evaluation-comparison.v1.json
Publication: json-schema
Defined fields: 17
Artifact bindings: 0

Why this contract exists

Architecture role

Defines a versioned interoperability boundary used by Workspai and contract-aware consumers.

Consumption boundary

• Workspai CLI
• Workspace Intelligence consumers
• IDEs and agents

Related commands

Commands are linked only when the runtime contract identifies a direct production relationship. Supporting contracts may be consumed without having a single producer command.

This is a supporting, capability, cache, compatibility, or consumer contract. It has no single direct producer command in the canonical Workspace Intelligence execution registry.

Produced artifacts

These durable files are emitted by registered commands and validated against this contract.

No canonical file artifact is registered for this contract. It may describe capabilities, embedded data, runtime exchange, or supporting state instead of a standalone report.

Field reference

Every declared schema property or published capability path is listed here. Required means the contract declares or publishes that field at the shown boundary.

Field path	Type	Required	Meaning	Constraints
$defs.run.runId	string	Yes	Defines the runid value at this contract boundary.	minLength: 1
$defs.run.status	enum	Yes	Defines the status value at this contract boundary.	enum: passed, failed, blocked, unknown
$defs.run.strategy	enum	Yes	Defines the strategy value at this contract boundary.	enum: full-corpus, grep, vector, graph, workspace-intelligence, custom
$defs.run.tokens	integer	Yes	Defines the tokens value at this contract boundary.	minimum: 0
$defs.run.verified	boolean	Yes	Defines the verified value at this contract boundary.	—
baseline	reference	Yes	Defines the baseline value at this contract boundary.	ref: #/$defs/run
comparableOutcome	boolean	Yes	Defines the comparableoutcome value at this contract boundary.	—
current	reference	Yes	Defines the current value at this contract boundary.	ref: #/$defs/run
delta	object	Yes	Defines the delta value at this contract boundary.	additionalProperties: false
delta.latencyMs	number	Yes	Defines the latencyms value at this contract boundary.	—
delta.modelCalls	number	Yes	Defines the modelcalls value at this contract boundary.	—
delta.reductionPercent	number \| null	Yes	Defines the reductionpercent value at this contract boundary.	—
delta.tokens	number	Yes	Defines the tokens value at this contract boundary.	—
delta.toolCalls	number	Yes	Defines the toolcalls value at this contract boundary.	—
generatedAt	string	Yes	Defines the generatedat value at this contract boundary.	format: date-time
schemaVersion	string	Yes	Defines the schemaversion value at this contract boundary.	const: "workspace-intelligence-evaluation-comparison.v1"
taskAligned	boolean	Yes	Defines the taskaligned value at this contract boundary.	—

Canonical machine-readable definition

The raw synchronized contract remains authoritative. The educational sections above are projections of this definition.

Show raw JSON

{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "https://getrapidkit.com/schemas/workspace-intelligence-evaluation-comparison.v1.json",
  "title": "Workspai Workspace Intelligence Evaluation Comparison",
  "description": "A task-aligned comparison between two provenance-aware Workspace Intelligence evaluation runs.",
  "type": "object",
  "required": [
    "schemaVersion",
    "generatedAt",
    "taskAligned",
    "current",
    "baseline",
    "delta",
    "comparableOutcome"
  ],
  "properties": {
    "schemaVersion": {
      "const": "workspace-intelligence-evaluation-comparison.v1"
    },
    "generatedAt": {
      "type": "string",
      "format": "date-time"
    },
    "taskAligned": {
      "type": "boolean"
    },
    "current": {
      "$ref": "#/$defs/run"
    },
    "baseline": {
      "$ref": "#/$defs/run"
    },
    "delta": {
      "type": "object",
      "required": [
        "tokens",
        "reductionPercent",
        "modelCalls",
        "toolCalls",
        "latencyMs"
      ],
      "properties": {
        "tokens": {
          "type": "number"
        },
        "reductionPercent": {
          "type": [
            "number",
            "null"
          ]
        },
        "modelCalls": {
          "type": "number"
        },
        "toolCalls": {
          "type": "number"
        },
        "latencyMs": {
          "type": "number"
        }
      },
      "additionalProperties": false
    },
    "comparableOutcome": {
      "type": "boolean"
    }
  },
  "$defs": {
    "run": {
      "type": "object",
      "required": [
        "runId",
        "strategy",
        "tokens",
        "verified",
        "status"
      ],
      "properties": {
        "runId": {
          "type": "string",
          "minLength": 1
        },
        "strategy": {
          "enum": [
            "full-corpus",
            "grep",
            "vector",
            "graph",
            "workspace-intelligence",
            "custom"
          ]
        },
        "tokens": {
          "type": "integer",
          "minimum": 0
        },
        "verified": {
          "type": "boolean"
        },
        "status": {
          "enum": [
            "passed",
            "failed",
            "blocked",
            "unknown"
          ]
        }
      },
      "additionalProperties": false
    }
  },
  "additionalProperties": false
}