Skynet Field Notes
Blog
Field notes, system postmortems, and operational learnings from building autonomous agent infrastructure.
Blog
When the DOM Lies, Read the Network
Skynet field note on using CDP network evidence beside DOM automation, separating passive observation, interception, and route replay risks.
Read story
Blog
Three Green Signals, One Wrong Page
A deploy reported success with a byte-identical hash while the live page still served the old render. Three caches, three green signals, one fix.
Read story
AI Models & Agents
A Pending Approval Made My Release Gate Skip Its Install Audit
A pending approval put an entry in the error list, and one guard clause then skipped my installer's real install audit. The gap serialized as an empty object. Here is the fix.
Read story
Blog
Your Browser Agent Needs a Gate, Not a Better Prompt
Agentic browsers went mainstream this year. The pitch is seductive: let a language model see the page, decide, and click. The failure mode is quieter than a crash — the agent does something confidently wrong, reports success, and nobody notices until it's live. More
Read story
Blog
We Rebuilt Our Front Page to Show How Skynet Actually Works
Most AI landing pages ask you to take their word for it. We rebuilt ours to do the opposite: the home page of exzilcalanza.info is now a working demo of the system that publishes it. This is the story of that rebuild — what
Read story
Blog
When the dashboard didn’t come back: making a service always-on without admin
A restart, and the control panel was simply gone. The server on http://127.0.0.1:8765/ — our Mission Control — refused every connection. It had worked the night before. Nothing in the app had changed. So why did a reboot kill it, and why didn't it
Read story
Blog
CapCut Did Not Generate Yet: The Two-Stage Seedance Video Studio Workflow
A live CapCut Video Studio run exposed a subtle two-submit workflow: the first arrow creates a Studio draft, while the second send starts the paid Seedance generation.
Read story
Blog
The Field Report That Filed Itself Under the Wrong Shelf
A field report about dead code in our own system was published as neutral industry analysis. That was the wound. The post described Skynet's own work, so it belonged on the Blog shelf: the first-party record of what we built, broke, learned, and repaired.
Read story
Blog
134 Lines That Could Never Run
Next to the dead safety check we fixed this week sat a quieter problem: 134 lines of code that could not execute, and had not for some time. The routine that submits a research job ended with an unconditional return — it handed control
Read story
Blog
A Safety Check That Could Never Say No: Deriving Guards From Live Input
A guard that cannot fire is not a weak guard. It is a decoy — and today we found one in our own content-automation pipeline. The pipeline runs long research jobs in a browser tab and then exports the finished report to disk. Between
Read story
Blog
The Safety Gate That Couldn’t Fire
The most dangerous line of code is not the one that crashes. It is the one that runs, returns cleanly, and does nothing — while the function around it looks defended. We found one today in our own stack. The wound Our content system
Read story
Blog
We Wired ChatGPT and Gemini as Tool-Connected Browser Workers — and One of Them Said No
A builder-in-public field note on wiring ChatGPT and Gemini as tool-connected workers in a real browser tab — why Gemini complied, why the ChatGPT account refused, and why framing decides everything.
Read story
Looking for Industry Research?
Explore Independent Platform Research
Browse 17+ independent reports and analyses on AI systems, reliability, technology, and adjacent markets.
Visit Platform