Brief a campaign from one paragraph
Turn a one-paragraph campaign request into a full brief with audience, angle, primary platform, KPI target, and risk flags. Static context pins the brand book so the brief never drifts off voice.
Three pains every sales and BD team hits weekly. Each one is what your reps actually complain about, not what a feature page would call them.
Writers spend Friday afternoon rewriting generic AI drafts because nothing sounds like the brand, and the queue still ships late.
An off-brand asset reaches LinkedIn before the brand director sees it, and the takedown thread eats the next standup.
Six channels run on three calendars, the launch slips one Monday in five, and nobody can replay which version actually went live.
Every pipeline below is a shape you wire on the canvas using the crew and tools further down. Not a feature we ship for you, a pattern you configure.
Turn a one-paragraph campaign request into a full brief with audience, angle, primary platform, KPI target, and risk flags. Static context pins the brand book so the brief never drifts off voice.
Generate the weekly post set across X, LinkedIn, and the blog, each grounded in the campaign brief. The rag_retrieve tool reads the won-campaign corpus so Copywriter reuses phrasings that already converted.
Pull search-console queries, competitor outlines, and the product corpus into a structured draft with H2 plan, citations, and meta block. HITL gate on publish so the editor approves before the post hits CMS.
Slot drafts into the per-platform posting windows for X, LinkedIn, TikTok, and YouTube, with hashtag and thread structure. Every scheduled item stages for approval through pause_for_human before any publish tool runs.
BrandArchitect checks palette, typography, voice, banned phrases, competitor mentions, and visual hierarchy on every asset. The run blocks at a HITL gate when any checklist item fails, so nothing off-brand leaves the tenant.
Take call notes, win-loss inputs, and product usage data into a structured customer story draft with quote candidates, metric callouts, and a meme companion. Cross-run memory carries prior interview answers so the second pass is not starting from zero.
Real personas from the marketing_team crew. Each ships with a tuned system prompt and a default tool allowlist. Swap models per persona on the canvas.
Designs the video and post concept, writes the hook, structures the shot list, and times every transition so the script lands on platform.
Writes the headline, three A/B variants, body copy, and CTA in a direct-response voice trained on the Ogilvy and Halbert tradition.
Designs the funnel, the viral loop, the seed list of accounts and communities, and the paid amplification plan against a target K factor.
Checks every asset against the brand brief on palette, typography, voice, and visual hierarchy, and surfaces inferred assumptions when guidelines are missing.
Turns the campaign into per-platform copy, hashtag strategy, posting windows, and thread or carousel structure for X, LinkedIn, TikTok, and YouTube.
Picks the canonical meme template that fits the angle, writes original captions inside the slot count, and renders one image per turn.
Every tool below is a real shared tool from the Melaya bundle. Allowlist per agent; HITL-gate the writes; revoke any of them in one click.
Load the brand book, tone guide, banned phrases, won-campaign corpus, and customer stories into the per-workflow store. Writes happen at build time, no HITL gate needed because nothing leaves your tenant.
Pull competitor posts, product launch coverage, and category news for ContentStrategist briefs. Read only, scoped per agent to the allowlist you set.
Search LinkedIn for the right surfaces, draft posts, and stage company-page publishes. linkedin_create_post and linkedin_send_message are HITL by default for this crew.
Mine X for trend angles and stage thread posts to the brand handle. twitter_post is HITL by default so the engineer can pull a draft before it ships.
Let MemeArtist look up the canonical template that matches an angle and render one image per turn. The generated file is HITL-staged in the run output before any social post tool can attach it.
Generate hero images, describe uploaded references for brand QA, and convert long-form posts to PDF for sales enablement. openai_image_generate is HITL-gated when the run feeds a publish tool.
Mine prior brand videos and competitor channels for hook patterns and structure. Read only, no publish path from this bundle.
General research, structured fetch, and Slack handoff for the approver. slack_post_text is HITL by default, and pause_for_human is the explicit approval node on the canvas.
Every pipeline ships with three layers of knowledge access. Mix and match per agent on the canvas. No shared vector space with another tenant, no surprise reads, no opaque retrieval.
includeContextPer-pipeline documents appended to specific agents' input on every run. The ICP brief, playbook, pricing sheet, or won-deal email corpus. Whatever needs to be there before the agent thinks. You pick which personas get which docs.
rag_retrieveA scoped tool granted per-agent. When the agent decides it needs more depth, it queries the workflow's vector store on demand. Same knowledge base as Static context, accessed only when the model asks for it.
pipeline_memoryPipeline-level state that carries from one run to the next. Yesterday's research is in scope for today's follow-up. The crew remembers what it already prospected, what got approved, what was sent. The audit log is the second-order knowledge base.
Sandbox tier is free with no card. Join the waitlist and we will email you the moment a slot opens.
← Back to every use case