Recipes for ContextuAI Solo crews. Assemble specialists from 96 prebuilt agents — or the unlimited custom agents you create — orchestrate them in Sequential, Parallel, or Pipeline mode, ground them in your Knowledge Base, and bind them to channels — with approvals and schedules.
Choose the orchestration that fits the job. Every mode can be grounded, bound to channels, and scheduled.
Agents work one after another, each building on the previous output — great for structured, staged deliverables.
Agents work simultaneously on the same brief. Merge or review each output — ideal for multi-channel and multi-language.
Each agent's output becomes the next stage's input — a clean assembly line from research to finished artifact.
Attach a Knowledge Base, bind channels, choose a model, add approvals, and schedule the whole crew on a cron.
Build a crew, pick an execution mode, and assign specialist agents.
Search 96 prebuilt agents (or create your own — custom agents are unlimited), add specialists, and choose Sequential, Parallel, or Pipeline.
Attach a Knowledge Base, choose a model, bind channels, and set approvals.
Run once with an input, or set a cron so the crew works on its own.
Publish 3 SEO-optimized posts a week without writing from scratch.
Features: Crew (Pipeline) + SEO Specialist, Content Strategist, Copywriter agents + Knowledge Base (brand guidelines) + Blog channel + M/W/F cron + Approvals
Setup: Upload brand guidelines as a KB. Pipeline crew: SEO Specialist picks keywords → Content Strategist outlines → Copywriter drafts (KB-informed). Bind Ghost/WordPress. Cron Mon/Wed/Fri 8 AM. Approve before publish.
PipelineKnowledge BaseScheduledProduce landing copy, an email sequence, social posts, and a press release in one workflow.
Features: Crew (Parallel) + Copywriter, Email Campaign Builder, Social Media Manager, PR Manager agents + LinkedIn, Twitter, Email channels + Approvals
Setup: Parallel crew with four agents, each producing their channel simultaneously. Bind LinkedIn, Twitter, Email. Run once with your launch brief. Review per channel in the approval queue, then publish.
ParallelMulti-ChannelApprovalsReview a diff from three angles — quality, bugs, and test coverage — before you ship.
Features: Coder Mode + Sequential crew + Code Reviewer, Bug Analyzer, Test Writer agents + project Knowledge Base
Setup: In Coder Mode, index the codebase as a KB. Sequential crew: Code Reviewer (quality) → Bug Analyzer (defects) → Test Writer (coverage gaps). Paste your diff and run.
Coder ModeSequentialKnowledge BaseAdapt English content into Spanish, French, and German while keeping brand voice.
Features: Crew (Parallel) + 3 custom language agents + Knowledge Base (brand voice + glossary) + Email or Slack output
Setup: Create 3 custom agents with per-language cultural prompts. Upload your glossary as a KB. Run a parallel crew — all three translate at once. Review in the approval queue, then push to localized channels.
ParallelCustom AgentsKnowledge BaseEvaluate a new market with landscape, regulation, TAM/SAM/SOM, and a go-to-market call.
Features: Crew (Pipeline) + Market Researcher, Competitor Intelligence Analyst, Compliance Officer, Growth Strategist agents + Web Research persona + PDF output
Setup: Pipeline crew: Market Researcher collects data via web search → Competitor Intelligence Analyst maps rivals → Compliance Officer flags risk → Growth Strategist synthesizes GTM. Export to PDF.
PipelineWeb ResearchPDFPost daily across LinkedIn, Twitter, and Instagram without hours of content work.
Features: Crew (Sequential) + Content Strategist, Copywriter, Social Media Manager agents + 3 channels + Daily cron + Approvals
Setup: Sequential crew: Content Strategist researches trends → Copywriter drafts platform posts → Social Media Manager adapts tone per platform. Bind all 3 channels. Daily cron 9 AM. Approve each post.
Sequential3 ChannelsDaily CronCrews run in Sequential (agents work one after another, each building on the last), Parallel (agents work simultaneously and you merge outputs), and Pipeline (each agent's output feeds the next stage) modes.
Yes. A crew can be grounded in a Knowledge Base for retrieval, bound to distribution channels for publishing or inbound replies, gated by approvals, and scheduled on a cron.
Start from 96 prebuilt agents, but you're not capped there — create unlimited custom agents of your own. Add as many specialists as the task needs and choose the model each crew uses. The 96 is a starting point, not a limit.
A crew is the team of agents and its execution mode. An automation is how you run it — chained via @mentions or fired on a schedule. You'll often build a crew, then automate it.
Download ContextuAI Solo for free and put a team of agents to work in minutes.