The More I Wrote in AGENTS.md, the Less Got Followed — Measuring Adherence and Cutting Rules
The rules in my AGENTS.md were being ignored — not from precedence conflicts or load failures, just plain ignored. Here is how I turned rules into checkable predicates, measured adherence over three weeks, and cut the file in half.
Three Ways to Hand 4,000 Lines of Logs to an Agent — Paste, .txt Attachment, or @ Reference
v2.3.0 added plain-text attachments, which means there are now three ways to hand a long log to an agent — and a new question about which one to pick. Here is how I trim, measure, and decide, with the scripts I actually run.
When to Hand Your Agent the Next Instruction: Waiting, Interrupting, and Queuing, Measured
Antigravity v2.3.0 added message queuing and Send Now. I measured waiting, interrupting, and queuing against the same yardstick, found that 41% of queued instructions arrive stale, and cut rework from 22% to 9% with a twenty-line stamp.
When Your Agent Commits a .bak File: Why Fix-Tool Artifacts End Up in Git
Backup files like .bak and .orig slip into commits after an agent runs a --fix tool. Here are the reproduction conditions, the real root cause, and three fixes: narrowing the staged scope, wrapping the fixer, and adding a pre-commit extension gate.
Quarantining the Dependencies Your Agent Adds, Before They Install
When an agent adds a dependency overnight, nobody reviews the lifecycle scripts that run at install time. Here is how I turned the default off and built a quarantine score to let the safe ones through.
When an Unattended Run Finds Its Working Folder Owned by Someone Else
A schedule that had been green for weeks failed one morning because it could not write .git/config. The cause was ownership drift on a reused working folder. Here is how to tell apart the read-but-not-write trap, probe writability for real, and fall back to a folder you can always write.
The File Is Right There in ls, and Your Agent Still Can't Open It
The agent says the file does not exist. Your terminal says it does. After three days of blaming cloud sync, the answer turned out to be that one voiced consonant mark was never a single character. Detection script and a three-layer gate included.
Protecting Screenshot Tests for AdMob Screens from Ad Nondeterminism
Screens with ads turn red on every screenshot run, and eventually nobody reviews the diffs. Here is a design that seals off AdMob banner nondeterminism and leaves only real layout breaks in your checks, with Compose code and Antigravity-driven diff triage.
When the Deploy Was Green but Users Still Saw the Old Build: Field Notes on a Gate That Verifies Your Shipped Commit Reached the Edge
When a deploy reports success but production keeps serving the old build, a post-deploy gate that verifies your shipped commit actually reached the edge via a build stamp closes the gap. Field notes with real operational numbers.
Routing Between Local Gemma and Cloud Gemini 3.5 Flash by How Easily You Can Verify the Output
When I split local and cloud work by whether it was sensitive, or by which was faster, my decisions wobbled every time. The axis that finally held was different: if the output is wrong, can I catch it cheaply and undo it cheaply? Here is a router that chooses a model from verifiability and recovery cost, with working code and a measurement ledger.
I Changed One Line in AGENTS.md and Broke the Agent — Regression-Testing the Instructions Themselves
Every time you edit the instructions you hand an agent — AGENTS.md, skill text — some other task quietly breaks. Here is a regression harness whose subject is the instructions, not the code: fixed tasks, deterministic assertions, and a way to wire it into unattended runs.
Passing API Keys to Agents Safely: Runtime Env Injection and Log Redaction
How to hand secrets to an Antigravity agent without leaving them in your repo or your logs, using runtime environment injection and output masking.