How did we unlock a fully local, cached, reproducible build and release CI/CD that cross compiles Go CGO binaries? With @dagger.io and @ziglang.bsky.social of course!
johncodes.com/archive/2026...
Open to any an all feedback! This is a legitimate project with real use cases for telemetry, auditability, and securing agents like claude code. So not certain what's "faux" about it.
Screenshots are a good idea although not certain what we'd even show! It's a data pipeline and context utility: you want to see server logs?
Support HTTPS_PROXY is abit of an anti-pattern because the proxy is just a reverse proxy to the upstream provider: we support CC through ANTHROPIC_BASE_URL
📼 Introducing tapes: transparent telemetry for your AI agents
Last week, we open sourced tapes, a new tool for agentic telemetry. We built this to help you understand the "what", "why", and "how" of your agents. It also requires zero code changes!!
johncodes.com/archive/2026...
Check out the release for the full details: and thank you to everyone who's thrown us a ⭐️ and our amazing first contributors!!
github.com/papercompute...
Download the latest today:
"curl -fsSL download.tapes.dev/install | bash"
Still very early, but you can run it today off `v0.0.1-rc.3` with: "tapes deck" and "tapes deck --web" for the local server web view
On the roadmap for tapes:
* deeper integration with more LLM providers (some people already asking for Google Vertex support!)
* emitting Open telemetry client side genAI metrics
* fully supporting agent-trace.dev
* TUI session search capabilities
📼 We've shipped a TUI for tapes!! Want to understand your previous agent sessions? Want to see success rates? Want to quickly grok token costs? We got it!
Check out the repo, throw us a star, and save open source ⭐️ github.com/papercompute...
Next up on our roadmap is partnering with Cursor to help tackle Agent-Trace: tracking the metrics and agent stats from AI coding assistance tools in our storage layer and search capabilities.
agent-trace.dev
$ tapes checkout <hash>
lets you do context check-pointing and retry. Using a hash from the session graph, you'll be able to then use "$ tapes chat" to time travel to a point-in-conversation and retry it.
$ tapes search "query"
lets you search back on previous conversation threads and sessions: since we use content addressing, you'll see the hash for that point in the conversation turn.
$ tapes chat
is our experimental chat client: this lets you see the full end to end experience since it uses the proxy by default and loads the point-in-conversation checkout state.
$ tapes serve
brings up 1) the proxy and 2) the api server.
The proxy connects to your configured LLM provider and transparently captures events and messages.
The API server makes commands like "search" and "checkout" work
How's it work?
Tapes has a number of components that enable it to have huge scale:
* a storage driver (sqlite locally by default)
* an LLM provider (ollama locally by default)
* an embedding provider (ollama locally by default)
* a vector store driver (sqlite-vec by default)
As @bizza.pizza would put it: tapes "is long term care for your agent sessions".
This is how we build the agentic platforms of the future with durable audit trails, memory, secure workflows, infrastructure, and strong loopback capabilities.
📼 Introducing tapes: transparent AI agentic telemetry. In every single AI session or workflow, there's an incredible amount of lost information on the "what", "why", and "how".
www.youtube.com/watch?v=ATeU...
Curious what I’ve been up to? Check it out: 🦞
Robin saying "Ingress nginx" and batman slapping him saying "No"
I'm once again reviving this meme template for the Kubernetes community
Joint statement from Kubernetes Steering and Security Response Committees: In March 2026, Kubernetes will retire Ingress NGINX, a piece of critical infrastructure for about half of cloud native environments. This is an emergency. Please pay attention.
kubernetes.io/blog/2026/01...
+1 It’s not just kubernetes nginx ingress: it’s the whole ecosystem. Open source program offices have been absolutely gutted and open source funding has dried up.
It’s a travesty that it takes a monumental project dying on the vine for people to wake up. The OSS maintenance tsunami is coming.
get in loser we’re building a new internet
#atproto
Ok, but what do you mean Claude Code won't respond to my obscure star trek references 😭
Still, over 10 years latter, the work by Onsi on Ginkgo and Gomega are having a huge impact on my dev flow. Providing really rich and expressive BDD testing in Go is incredible, especially in the age of AI assisted code editing
onsi.github.io/ginkgo
If you bring down the service’s infrastructure, so be it? 😂 page sizes are not a UI choice: they’re often an infrastructure constraint and guardrail.
