Continuous Runtime Feedback
AI is writing the code.
Who’s watching the runtime?
AI writes the code, configures the infra, triggers the deploy. When something breaks — you’re starting from zero. No context, no mental model, no idea where to look.
Dstl8 is the feedback loop between AI-generated code and runtime reality.
brew install control-theory/dstl8/dstl8
Works with your stack
Kubernetes
OpenTelemetry
CloudWatch
Claude Code
Cursor
Codex
Copilot
Datadog
Vercel
Supabase
Railway
AWS
Runtime platforms and AI coding tools — connected in minutes. See all integrations →
The Runtime Context Gap
The faster AI ships code,
the less anyone understands what it’s doing.
01
Shipping complexity faster than anyone can follow
Every prompt ships a feature, not a function. More features, more moving parts, more emergent interactions. System complexity is outpacing understanding — and nobody in the room wrote the code.
# 7 PRs merged by Cursor in the last hour
2 new API routes, 1 dependency upgrade
● team members who reviewed
the runtime behavior:
— none —
02
Deployment chains you can set up but can’t debug
Vercel, Supabase, Railway — quick to configure, impossible to correlate when something breaks across them. Each platform has its own logs, its own schema, its own propagation delay. The stack is a black box assembled from other black boxes.
# same incident, three different surfaces
vercel: edge function timeout 11:00:01
supabase: RLS policy denial 11:00:04
railway: connection pool exhaust 11:00:07
→ no cross-source correlation out of the box
03
Every failure is a rabbit hole you didn’t earn
The developer didn’t write the code, didn’t configure the infra, didn’t choose the dependency. They have no mental model to start from. The knowledge tax is enormous — you need to know what to look for before you can look for it.
# monday morning incident response
1. open Vercel dashboard (which project?)
2. find the failing function (which route?)
3. read the error (what does this mean?)
4. check Supabase logs (different timestamp)
5. check Railway logs (different schema)
6. give up and ask the person who prompted it
→ they don’t remember either
— none —
This is the runtime context gap. It widens with every AI improvement — and it’s where developer progress goes to die.
// See how Dstl8 closes the gap.
Closing the Runtime Feedback Loop
Dstl8
Always-On
Continuous Runtime Distillation
Dstl8 watches your runtime continuously — catching regressions, unexpected behavior, and unknown-unknowns before they snowball. Not periodic checks. Not dashboards you have to stare at. Always-on, proactive feedback that keeps pace with AI-generated code.
Cross-platform
Full-Stack Runtime Intelligence
Supabase, Vercel, Railway, app logs, k8s events, CloudWatch, OTLP streams — Dstl8 correlates issues across your entire deployment chain, not siloed by source. Powered by Möbius AI for root cause, impact assessment, and fix recommendations.
Developer flow state
In-Context Runtime Feedback
Dstl8 feeds runtime insights directly into Claude Code, Cursor, Codex, and your existing tools. No context switching. No knowledge tax. No rabbit holes. And it compounds — every signal makes the next diagnosis faster.
14 days free · No credit card · Full platform access
Start Here
See what’s actually happening.
Connect your deployment chain. Surface emergent patterns. Get root cause analysis with fix recommendations — right in your editor.
↻ Intelligence that compounds — every runtime signal makes the next one sharper.
Dstl8 — Supabase runtime analysis
Open Source
Not ready for Dstl8? Start with Gonzo.
Free, open source log analysis TUI. Real-time charts, pattern detection, AI-powered insights — right in your terminal. No account, no config.
brew install gonzo
“
I decided to give it a shot. It’s really nice! One of the things I always loved about Datadog’s log analysis tool was its ability to surface log patterns.
— Senior SRE · Series B SaaS
Why controltheory
Not another dashboard. A feedback loop.
Finds
Correlation
Debugging
Collection
AI code
Dev flow
Over time
Traditional monitoring
What you configured
Within one platform
Dashboard archaeology
Collect, pay for everything
Not designed for it
Context switch to dashboard
Same dashboards forever
ControlTheory
Unknown-unknowns
Across your deployment chain
Root cause with evidence
Distill only what you need
Built for it
Insights in your terminal
Intelligence that compounds
Get started
Install & Configure Dstl8 in Under 2 Minutes.
Try the Dstl8 CLI and TUI for continuous runtime feedback. Install it, add sources, connect the MCP server into Claude Code, and more.
brew install control-theory/dstl8/dstl8
dstl8 signup
curl -fsSL https://install.dstl8.ai/script/dstl8-cli | sh
npx dstl8
nix run github:control-theory/dstl8
Download from GitHub Releases
Quick Start
# 1. Install the CLI
brew install control-theory/dstl8/dstl8
# 2. Create a Dstl8 account (or `dstl8 login` if you already have one)
dstl8 signup
# 3. Add a source so logs flow in
dstl8 sources add vercel
# 4. Connect your AI agent, auto-detects MCP-compatible clients on your machine and configures them
dstl8 install -all
dstl8 install claude-code
Add Sources
# Add Sources
dstl8 sources add kubernetes
dstl8 sources add cloudwatch
dstl8 sources add vercel
dstl8 sources add supabase
dstl8 sources add otlp
dstl8 sources add github