IntraCord

Create a reusable tool the agent can invoke during calls. The request schema is the same `CreateToolRequest` used by the REST API and generated SDKs. Use it to create HTTP API, end-call, transfer-call, calculator, or MCP-server tools. For authenticated HTTP or MCP tools, reference an existing `credential_uuid` from `list_credentials`; users create credential secrets in the UI, and this flow only stores the UUID reference. For MCP tools, the server best-effort discovers the remote tool catalog and caches it in `definition.config.discovered_tools`. On success, returns `created: true` and the new `tool_uuid`; use that UUID in workflow node `tool_uuids`. On failure, returns `created: false`, a machine-readable `error_code`, and a human-readable `error`. Possible `error_code` values: - `validation_error` — the request failed schema validation. - `credential_not_found` — a supplied credential_uuid is not in this organization; ask the user to create/select it in the UI first. - `organization_required` — the API key user has no selected organization. - `hosted_interactive_admin_required` — hosted tool authoring requires interactive owner/admin access; API-key-backed MCP is read-only here. - `create_failed` — unexpected persistence or backend failure; retry once, then surface the error.

Parse SDK TypeScript and create a new published workflow. `code` is TypeScript source using `@IntraCord/sdk`. The workflow name comes from `new Workflow({ name: "..." })` — it is required. Example code: import { Workflow } from "@IntraCord/sdk"; import { startCall, endCall } from "@IntraCord/sdk/typed"; const wf = new Workflow({ name: "lead_qualification" }); const greeting = wf.addTyped(startCall({ name: "Greeting", prompt: "Hi!" })); const done = wf.addTyped(endCall({ name: "Done", prompt: "Bye." })); wf.edge(greeting, done, { label: "done", condition: "conversation complete" }); On success the new workflow is published as version 1. Use `save_workflow(workflow_id, code)` for subsequent edits — those go to a draft. On failure the result has `created: false`, a machine-readable `error_code`, and a human-readable `error` (with file:line:column where the problem is locatable). Resubmit the full corrected source — patches are not accepted. Possible `error_code` values: - `parse_error` — disallowed construct or malformed TypeScript. - `validation_error` — node data failed spec validation (unknown field, missing required, wrong type, option out of range). - `schema_validation` — wire-format (DTO) rejection; rare. - `graph_validation` — structural rule broken (e.g. no start node, unreachable node, edge to/from the wrong node type). - `missing_name` — `new Workflow({ name })` is absent or empty; the name is required and there is no prior workflow to fall back to. - `trigger_path_conflict` — a trigger node's path is already used by another workflow in this organization; rename it and resubmit. - `bridge_error` — internal/transient; retry once, then surface it. - `hosted_interactive_admin_required` — hosted workflow authoring needs interactive owner/admin access; API-key-backed MCP cannot create it.

Fetch the speaker-labeled transcript for a single call. Returns the transcript as a plain-text string. Returns an empty string only when no transcript was ever recorded for this call (no stored transcript file). A storage/download failure raises an error instead, so callers can tell "no transcript" apart from "couldn't fetch it".

Fetch the authoring schema for a node type: each property's name, type, default, requiredness, enum options, validation bounds, and LLM-readable description, plus worked examples and graph constraints. UI-only metadata (display labels, placeholders, conditional visibility rules, renderer hints) is intentionally omitted — set only the fields you need. Use the property `description`/`llm_hint` and the `examples` list to understand semantics; types alone are not enough.

Fetch staged voice-prompting guidance for authoring IntraCord workflows. Call this BEFORE composing or revising any prompt field on a node. The guide is the authoritative source for prompt-authoring craft (global guidelines, turn-taking, tool calls, success criteria, guardrails); product-mechanics questions (how a node type works at runtime) belong in `search_docs` / `read_doc`.

Fetch a single agent by id, projected into the SDK code view. Output shape: {"id": int, "name": str, "status": str, "version": "draft" | "published" | "legacy", "version_number": int | None, "code": "<TS source>"}

Return the workflow as SDK TypeScript code the LLM can edit. Output shape: {"code": "<TS source>", "workflow_id": int, "version": "draft" | "published" | "legacy"} The LLM edits `code`, then calls `save_workflow(workflow_id, code)`.

List completed voice calls for a workflow, newest first. Returns id, call_type, duration_seconds, disposition, and created_at for each call. The id is the call identifier — pass it straight to get_call_transcript. Excludes text-chat sessions and incomplete runs.

List external credentials available for webhook auth and pre-call fetch. Returns each credential's `credential_uuid` (use this in node `credential_uuid` / `pre_call_fetch_credential_uuid` properties), `name`, `description`, and `credential_type`.

Browse the IntraCord docs hierarchy before reading a page in full. ``path`` addresses navigation sections exposed by this tool. Page paths returned by ``search_docs`` and ``read_doc`` are stable document identifiers, for example ``voice-agent/tools/mcp-tool``. The ``url`` is the public route.