Get agent context pack
get_agent_context_packUse for research questions that need a budgeted, explainable context bundle.
When to use it
Use for research questions that need a budgeted, explainable context bundle. Select essentials, standard, or deep and pass task, surface, session, and active object hints when known.
Orient first, then choose the smallest task-specific tool.
Workspace availability, recent objects, tool routing, or query-focused context.
Avoid using it when
You already have the exact source, draft, post, artifact, or portfolio id.
Advanced input contract
Agent hosts use this JSON Schema to validate the exact input shape.
View full registry notes
Return an authorized Prism context pack. legacy_v1 retains compatibility fields; compact_v2 returns one bounded, versioned envelope without raw traces or duplicated content. Use when: Use for research questions that need a budgeted, explainable context bundle. Select essentials, standard, or deep and pass task, surface, session, and active object hints when kno...
View JSON Schema
{
"type": "object",
"properties": {
"query": {
"type": "string",
"minLength": 1,
"maxLength": 240
},
"folderId": {
"type": "string",
"format": "uuid"
},
"limit": {
"type": "integer",
"minimum": 1,
"maximum": 20,
"default": 8
},
"task": {
"type": "string",
"minLength": 1,
"maxLength": 500
},
"surface": {
"type": "string",
"enum": [
"mcp",
"rest",
"chat",
"mobile",
"system"
],
"default": "mcp"
},
"sessionId": {
"type": "string",
"minLength": 1,
"maxLength": 200
},
"activeObjectIds": {
"type": "array",
"maxItems": 20,
"items": {
"type": "string",
"format": "uuid"
}
},
"profile": {
"type": "string",
"enum": [
"essentials",
"standard",
"deep"
],
"default": "standard"
},
"maxTokens": {
"type": "integer",
"minimum": 256,
"maximum": 12000,
"description": "Maximum context token budget. When provided, this takes precedence over legacy limit."
},
"intent": {
"type": "string",
"enum": [
"resume",
"verify",
"challenge",
"changes",
"research"
]
},
"since": {
"type": "string",
"format": "date-time"
},
"format": {
"type": "string",
"enum": [
"compact_v2",
"legacy_v1"
],
"default": "legacy_v1"
},
"asOf": {
"type": "string",
"format": "date-time",
"description": "Historical context reconstruction is currently unavailable and returns an explicit error. Use get_belief_timeline for recorded changes; omit asOf for current context."
},
"includeTrackRecord": {
"type": "boolean",
"default": true,
"description": "Attach the user's resolved-hypothesis track record (hit rate, Brier score, calibration hints) so stated confidence can be weighed against history."
}
},
"additionalProperties": false
}