Agents / TypeScript SDK

Get agent skill loadout

Name: Pipeline
Price: 99 USD

Resolve the effective role and task skill loadout for an external coding agent, including workspace overlays, provider SOPs, versions, hashes, required context, and MCP resource URIs. Never invokes hosted inference.

pipeline.getAgentSkillLoadoutDeterministic operation

Parameters

workspaceIdOptional · string: Optional with API key auth, inferred from the key.
agentRoleRequired · string: See the input schema for this field.scout · researcher · icp · strategist · writer · campaign_builder · closer · sdr · coach · curator · drafter · publisher · prospector · watcher · commenter
taskTypeRequired · string: See the input schema for this field.campaign_strategy · campaign_table_design · company_list_build · people_fanout · account_map · company_research · person_research · personalization_research · signal_research · icp_qualification · linkedin_connection_note · linkedin_dm · linkedin_followup · reply_classification · reply_draft · campaign_analysis · experiment_design · content_ideation · content_draft
objectiveOptional · string: See the input schema for this field.

Request example

await pipeline.getAgentSkillLoadout({
  "agentRole": "scout",
  "taskType": "campaign_strategy"
})

Full input schema

Input schema

{
  "type": "object",
  "properties": {
    "workspaceId": {
      "type": "string",
      "description": "Optional with API key auth, inferred from the key."
    },
    "agentRole": {
      "type": "string",
      "enum": [
        "scout",
        "researcher",
        "icp",
        "strategist",
        "writer",
        "campaign_builder",
        "closer",
        "sdr",
        "coach",
        "curator",
        "drafter",
        "publisher",
        "prospector",
        "watcher",
        "commenter"
      ]
    },
    "taskType": {
      "type": "string",
      "enum": [
        "campaign_strategy",
        "campaign_table_design",
        "company_list_build",
        "people_fanout",
        "account_map",
        "company_research",
        "person_research",
        "personalization_research",
        "signal_research",
        "icp_qualification",
        "linkedin_connection_note",
        "linkedin_dm",
        "linkedin_followup",
        "reply_classification",
        "reply_draft",
        "campaign_analysis",
        "experiment_design",
        "content_ideation",
        "content_draft"
      ]
    },
    "objective": {
      "type": "string"
    }
  },
  "required": [
    "agentRole",
    "taskType"
  ]
}

See this operation inCLI MCP REST API