The Autonomous Pipeline
Every post, thread, clip, and ad creative goes through ten stages. From live platform intelligence to a live thread shaped for every connected feed, scheduled into the right send window, with paid-ad creative ready for rotation. No human in the loop. About 2 minutes end to end.
The same pipeline also powers networkr.dev (SEO articles) and outboxr.dev (email + outreach). One engine, three surfaces, one /connect.
Intelligence Gathering
Pulls real-time signals from multiple sources: platform engagement trends, breaking conversations in your category, trending repositories, ship signals from connected repos, industry forecasts. Multi-region support so content matches your audience's timezone. Today's signals, not last month's training data.
Topic Selection
The system maintains a topic queue per account, filled weekly by the suite's topic engine. If the queue has pending topics discovered from real audience trends or commit signals, it pulls the highest priority. Otherwise, the LLM analyzes today's intelligence against your account's themes and picks an angle. Built-in deduplication across the whole suite so nothing repeats between viralr posts, networkr articles, and outboxr newsletters.
Competitive Shaping
Searches platform-native conversations for the chosen topic. Pulls top-performing posts on the same angle, per platform. The generation stage knows what's already landing, and is told to write something sharper and more specific.
Context Assembly
Gathers everything the system knows about your account: featured content, brand tone, site profile, audience signals, banned words, past engagement winners. This context shapes each post so it reads like it was written by someone who knows your product and voice.
Cross-Platform Reshape
One angle, five forms. X gets a threaded hook. LinkedIn gets a narrative. Threads gets a single scroll-stop. Bluesky gets a quieter version. Mastodon gets a long-form take. Same message, different feed. Every reshape runs through its own voice and length gates.
Post Generation
The LLM writes the full post in your account's voice. Persona, tone, structure, and content rules enforced. Prompt includes platform-native shaping, exposure manifest, past winners, and banned phrases. The system doesn't generate posts. It generates posts that land.
Hook / Body / CTA
The post is shaped into a scroll-stop hook, a body that delivers what the hook promised, and a CTA optimized for signup, star, follow, or share. The hook/body consistency gate blocks ragebait openers that contradict what the body says.
Media + Clip
Short-form video clips generated from the post body where the platform rewards them (TikTok, Shorts, Reels). Branded visuals, carousels, and OG cards generated per platform. Each account gets a distinct visual style so no two accounts look alike.
Schedule + Post
The scheduler picks the peak-engagement window per platform per audience. The post goes out through your connected accounts. Paid-ad variants of winning posts are queued for the viralr-ads rotation if you've opted in.
Suite Propagation
Best-performing threads seed networkr's next SEO article and outboxr's newsletter lineup. Engagement signals update the topic queue across all three surfaces. Every new post makes the whole suite slightly smarter.
What runs in the background
The pipeline is the visible part. Behind it, four autonomous systems run continuously to keep the accounts healthy, the topic strategy sharp, and the audience growing.
Topic Engine
Runs weekly. Researches platform-native trends and audience conversations for every theme on every account. Discovers angle gaps, rising questions, and ship-signal opportunities. Generates topic ideas and queues them for the pipeline. Your content strategy evolves with the feed.
Orchestrator
Runs daily. Self-monitoring system that checks pipeline health, platform reachability, post-to-engagement ratios, ad-creative performance, topic queue depth, and suite health. Generates actionable recommendations and flags issues before they become problems.
Safety Monitor
Runs continuously. Monitors every account for platform warnings, rate-limit patterns, spam signals, and policy drift. Accounts approaching a warning threshold pause that surface automatically before the platform acts. The suite protects itself.
Weekly Digest
Every Monday morning. Posts shipped, engagement movements with direction arrows, ad-creative performance, topic queue status, and recommendations. Delivered to your inbox. No login needed. No dashboard to check.
Builder-log mode (optional)
If you are building a product in public, Viralr turns your real shipped work into weekly thread-papers. First person, verifiable against commit SHAs, indexed as you build. Two ways to wire it up. Both use the same pipeline, the same quality gates, and the same immutable archive. Pick the one that matches how much control you want.
Fully automated
Connect GitHub once. Our pipeline pulls a week of commit metadata (SHAs, messages, dates, changed file paths), clusters by theme, extracts decisions from PR titles and ADRs, and writes the thread. You do nothing.
- ✓Read-only scope. Metadata only. Never code, never diffs, never file contents.
- ✓Works on private repos.
- ✓/private: in a commit message excludes that commit entirely.
- ✓Metadata purged after the thread publishes. Only the finished thread persists.
Your agent via API
No OAuth. npx viralr init drops .viralr/AGENT.md and a post-merge hook in your repo. Your own IDE agent reads git log locally, filters, clusters, and POSTs a structured narrative to our API. Only the narrative crosses the wire.
- ✓Code never leaves your machine. Architecturally impossible. The payload schema has no field for diffs.
- ✓Hook fires on merge, or run npx viralr report manually whenever.
- !Semi-automated. If your agent isn't available when the hook fires, no thread gets generated that week.
- !Silence on your side equals silence on the feed. We never fabricate a thread to hit cadence.
Both paths converge on the same pipeline, same prose guardrails, same lineage record, same immutable archive. Path A trades control for zero touch. Path B trades zero touch for full data sovereignty.
Voice System
Every account has a voice that controls how posts are written. Voice is separate from author. Voice is the style, author is the byline. The system enforces voice rules at generation time, including banned words, structure preferences, and tone constraints. The same voice profile shapes networkr's articles and outboxr's newsletters, so your presence sounds like one person across every surface.
Custom voices supported. Define your own persona, tone, banned words, structure rules, and preferred model.