Tool tree
gradeNot graded
get_workspaceAnswer
Get the current ONCE vendor-evaluation workspace, including budget, candidates, criteria, evidence, scores, uncertainty, recommendation, phase, and replay status. Call this when you need the current shared state before deciding what to change.
View tool JSON
{
"name": "get_workspace",
"kind": "answer",
"impl": "imperative",
"description": "Get the current ONCE vendor-evaluation workspace, including budget, candidates, criteria, evidence, scores, uncertainty, recommendation, phase, and replay status. Call this when you need the current shared state before deciding what to change.",
"inputSchema": {
"type": "object",
"properties": {},
"additionalProperties": false
}
}add_candidatesAction
Add one or more vendor candidates to the current collaboration workspace. Candidate identities become variable inputs when a routine is taught. Do not use this tool to change candidates after replay has started.
View tool JSON
{
"name": "add_candidates",
"kind": "act",
"impl": "imperative",
"description": "Add one or more vendor candidates to the current collaboration workspace. Candidate identities become variable inputs when a routine is taught. Do not use this tool to change candidates after replay has started.",
"inputSchema": {
"type": "object",
"properties": {
"candidates": {
"type": "array",
"minItems": 1,
"maxItems": 4,
"items": {
"type": "object",
"properties": {
"name": {
"type": "string",
"minLength": 1
}
},
"required": [
"name"
],
"additionalProperties": false
}
}
},
"required": [
"candidates"
],
"additionalProperties": false
}
}get_vendor_dossierAnswer
Read ONCE's deterministic demo dossier facts for one or more named vendors. Use these first-party facts as evidence inputs for the competition demo. Do not claim the dossier is live web research.
View tool JSON
{
"name": "get_vendor_dossier",
"kind": "answer",
"impl": "imperative",
"description": "Read ONCE's deterministic demo dossier facts for one or more named vendors. Use these first-party facts as evidence inputs for the competition demo. Do not claim the dossier is live web research.",
"inputSchema": {
"type": "object",
"properties": {
"vendorNames": {
"type": "array",
"minItems": 1,
"maxItems": 4,
"items": {
"type": "string",
"minLength": 1
}
}
},
"required": [
"vendorNames"
],
"additionalProperties": false
}
}get_replay_planAnswer
Get the active learned routine, its new input bindings, completed and pending steps, approval status, and the next allowed semantic work. Call this when the workspace is in replay mode.
View tool JSON
{
"name": "get_replay_plan",
"kind": "answer",
"impl": "imperative",
"description": "Get the active learned routine, its new input bindings, completed and pending steps, approval status, and the next allowed semantic work. Call this when the workspace is in replay mode.",
"inputSchema": {
"type": "object",
"properties": {},
"additionalProperties": false
}
}set_budgetAction
Set the evaluation budget in the current collaboration workspace. In taught routines, budget becomes a variable input rather than a fixed policy. This action is not allowed after an active replay has started.
View tool JSON
{
"name": "set_budget",
"kind": "act",
"impl": "imperative",
"description": "Set the evaluation budget in the current collaboration workspace. In taught routines, budget becomes a variable input rather than a fixed policy. This action is not allowed after an active replay has started.",
"inputSchema": {
"type": "object",
"properties": {
"amount": {
"type": "number",
"minimum": 0
},
"currency": {
"type": "string",
"enum": [
"USD"
]
}
},
"required": [
"amount",
"currency"
],
"additionalProperties": false
}
}add_criteriaAction
Add one or more evaluation criteria with priority and required status. Criteria are treated as reusable evaluation policy when a routine is taught.
View tool JSON
{
"name": "add_criteria",
"kind": "act",
"impl": "imperative",
"description": "Add one or more evaluation criteria with priority and required status. Criteria are treated as reusable evaluation policy when a routine is taught.",
"inputSchema": {
"type": "object",
"properties": {
"criteria": {
"type": "array",
"minItems": 1,
"maxItems": 6,
"items": {
"type": "object",
"properties": {
"name": {
"type": "string",
"minLength": 1
},
"description": {
"type": "string"
},
"priority": {
"type": "integer",
"minimum": 1
},
"required": {
"type": "boolean"
}
},
"required": [
"name",
"priority",
"required"
],
"additionalProperties": false
}
}
},
"required": [
"criteria"
],
"additionalProperties": false
}
}attach_evidenceAction
Attach evidence summaries to candidate/criterion pairs using source references from the ONCE demo dossiers. During replay, evidence is required for every candidate/criterion pair before scoring can complete. Evidence text is execution output and is not copied literally into a learned routine.
View tool JSON
{
"name": "attach_evidence",
"kind": "act",
"impl": "imperative",
"description": "Attach evidence summaries to candidate/criterion pairs using source references from the ONCE demo dossiers. During replay, evidence is required for every candidate/criterion pair before scoring can complete. Evidence text is execution output and is not copied literally into a learned routine.",
"inputSchema": {
"type": "object",
"properties": {
"items": {
"type": "array",
"minItems": 1,
"maxItems": 12,
"items": {
"type": "object",
"properties": {
"candidateId": {
"type": "string",
"minLength": 1
},
"criterionId": {
"type": "string",
"minLength": 1
},
"summary": {
"type": "string",
"minLength": 1
},
"sourceRef": {
"type": "string",
"minLength": 1
},
"confidence": {
"type": "string",
"enum": [
"high",
"medium",
"low"
]
}
},
"required": [
"candidateId",
"criterionId",
"summary",
"sourceRef",
"confidence"
],
"additionalProperties": false
}
}
},
"required": [
"items"
],
"additionalProperties": false
}
}set_scoresAction
Score one or more candidate/criterion pairs from 1 to 5 with a rationale. Use 1=does not meet, 2=materially below, 3=meets, 4=strong, 5=excellent. During replay, use current evidence and routine policy. A replay score requires evidence for the same pair. Score values are generated per run and are not copied from the taught session.
View tool JSON
{
"name": "set_scores",
"kind": "act",
"impl": "imperative",
"description": "Score one or more candidate/criterion pairs from 1 to 5 with a rationale. Use 1=does not meet, 2=materially below, 3=meets, 4=strong, 5=excellent. During replay, use current evidence and routine policy. A replay score requires evidence for the same pair. Score values are generated per run and are not copied from the taught session.",
"inputSchema": {
"type": "object",
"properties": {
"items": {
"type": "array",
"minItems": 1,
"maxItems": 12,
"items": {
"type": "object",
"properties": {
"candidateId": {
"type": "string",
"minLength": 1
},
"criterionId": {
"type": "string",
"minLength": 1
},
"score": {
"type": "integer",
"minimum": 1,
"maximum": 5
},
"rationale": {
"type": "string",
"minLength": 1
}
},
"required": [
"candidateId",
"criterionId",
"score",
"rationale"
],
"additionalProperties": false
}
}
},
"required": [
"items"
],
"additionalProperties": false
}
}flag_uncertaintyAction
Record an uncertainty that should remain visible to the human. Use this when the available evidence does not justify a confident conclusion. The literal uncertainty note is not copied into a routine.
View tool JSON
{
"name": "flag_uncertainty",
"kind": "act",
"impl": "imperative",
"description": "Record an uncertainty that should remain visible to the human. Use this when the available evidence does not justify a confident conclusion. The literal uncertainty note is not copied into a routine.",
"inputSchema": {
"type": "object",
"properties": {
"candidateId": {
"type": "string",
"minLength": 1
},
"criterionId": {
"type": "string",
"minLength": 1
},
"note": {
"type": "string",
"minLength": 1
}
},
"required": [
"note"
],
"additionalProperties": false
}
}set_recommendationAction
Set the current final recommended candidate and rationale. In replay mode, this tool is blocked until every required pre-recommendation step is complete and any required human approval has been explicitly granted.
View tool JSON
{
"name": "set_recommendation",
"kind": "act",
"impl": "imperative",
"description": "Set the current final recommended candidate and rationale. In replay mode, this tool is blocked until every required pre-recommendation step is complete and any required human approval has been explicitly granted.",
"inputSchema": {
"type": "object",
"properties": {
"candidateId": {
"type": "string",
"minLength": 1
},
"rationale": {
"type": "string",
"minLength": 1
}
},
"required": [
"candidateId",
"rationale"
],
"additionalProperties": false
}
}