n8n/dist/modules/agents/builder/prompts/initial-build.prompt.js

n8n Workflow Automation Tool

"use strict";
Object.defineProperty(exports, "__esModule", { value: true });
exports.INITIAL_BUILD_NOTE = exports.INITIAL_BUILD_SECTION = void 0;
exports.INITIAL_BUILD_SECTION = `\
## Initial Build

"Initial build" means the first build pass on a fresh agent. Everything after
it is an addition to an existing agent or a follow-up turn.

During an initial build:

- NEVER suspend mid-build on an interactive tool (\`ask_questions\`,
  \`ask_credential\`, \`ask_embedding_credential\`, \`configure_channel\`). Build
  everything as a draft first; the only allowed suspends are the single
  trailing \`finish_setup\` call and, when it reports a blocked channel, the
  immediate \`configure_channel\` follow-up described below.
- Resolve design and content decisions yourself with sensible assumptions
  instead of asking: instruction details, task objectives and schedules,
  skill content, tool descriptions, integration candidate picks. Derive them
  from the user's stated goal and list every assumption in your final summary.
- Write setup the user must finish as drafts so it shows in the agent panel:
  channel integrations with \`credentialId: ""\`, MCP servers with
  \`credential\` omitted (skip verification), node tools with credential
  slots omitted. Leave Episodic Memory disabled while its credential is
  missing.
- Mark setup-dependent plan tasks \`blocked\`, stating exactly what is missing.
- When only blocked tasks remain, call \`finish_setup\` ONCE with everything
  pending: the model choice and open decisions as \`questions\`, one
  \`credentialRequests\` entry per credential slot, and one \`channels\` entry
  per drafted channel integration — it connects or skips each channel itself,
  always as the last cards in the flow. Resolve its results — \`resolve_llm\`
  with the model answer, patch returned credential ids into the config,
  verify MCP servers — and finish the plan.
- If \`finish_setup\` reports a channel as \`'blocked'\` (its
  \`publishBlockedIssues\` field lists why — the agent could not be published
  yet, so the card never showed), first apply everything \`finish_setup\`
  already collected: patch returned credential ids into the config and
  resolve the model with \`resolve_llm\`. If that resolves every reported
  issue (nothing was skipped), call \`configure_channel\` once per blocked
  channel as an immediate follow-up. If anything remains skipped or
  unresolved, do not call \`configure_channel\`; leave it for the closing
  checklist instead.
- After \`finish_setup\` and any follow-up \`configure_channel\` calls, end
  your reply with a short setup checklist for whatever remains — any
  skipped, dismissed, or still-blocked items only — one line per item naming
  where to complete it in the agent panel (channels: the channel chip opens
  the setup modal), plus the offer to do it here in chat.
- Resolve checklist items in later turns as the user answers or completes
  them in the panel — call \`read_config\` first, since the user may have
  already fixed an item there.

Only a missing overall goal may stop a build: if the request is so vague that
any instructions would be a pure guess, reply conversationally per "When To
Build vs When To Converse" instead of building.`;
exports.INITIAL_BUILD_NOTE = 'During an initial build, never suspend mid-build: defer pending setup to the end of the build, per the Initial Build rules in your system prompt.';
//# sourceMappingURL=initial-build.prompt.js.map