All Articles
Contract-Test Every Tool Before You Submit or Automate an MCP Connector
A connector that works once in a chat can still break silently in an unattended job through misread response shapes or double-fired writes. Here is a small harness that machine-checks tool descriptions, response contracts, idempotency, and latency, with measured numbers.
A connector failed for two nights and I never noticed — instrumenting my solo setup after observability went to public beta
The week connector observability hit public beta, I realized my one-person operation had no view into errors or latency. Here is how I wrapped my MCP connector calls in a thin meter and started reviewing it weekly.
Four sites, one Claude bill: attributing spend per workspace to decide which pipeline to trim
When several projects share one Claude organization, the bill arrives as a single number that hides which pipeline is expensive. Field notes on splitting that spend per workspace with the Cost and Usage Report group_by, defensively parsing the results, and deciding what to trim by cost-effectiveness.
I Could Detect Failures — Then the Alerts Grew So Loud I Stopped Noticing
The moment I could detect silent job failures, my phone started buzzing too often to be useful. Here is how I collapsed alerts by action rather than event, added hysteresis and quiet hours, and built a three-step escalation so only the pages worth waking me survive.
When Claude API Suddenly Starts Returning 429 in Production — Field Notes on Measuring Rate-Limit Headroom from Headers and Throttling Before You Run Dry
Scrambling to add retries after a 429 is always a step behind. Claude API writes how much you have left into every response header. These are field notes on measuring that headroom continuously and throttling yourself before it runs out.
My background session sat at running all night — adding heartbeats to the agents view
The Claude Code agents view finally lets you see every background session at once, but a running badge is not proof of progress. Here is the external heartbeat layer I built to catch a silently stalled session in minutes instead of the next morning, with real numbers and the traps I hit.
Which Nightly Job Wrote This Commit? Threading a Correlation Key Through Claude Code's Readable Session Names
When you run unattended nightly jobs across several repositories, you lose track of which session produced which commit. Here is how I redesigned Claude Code's readable session names into a correlation key that ties commits, logs, and sessions into a single thread — with real numbers.
When Instructions Fade in Long Conversations: Measuring Adherence and Pulling It Back
System-prompt instructions quietly lose their grip as a conversation grows. This piece turns that drift into a number you can track, then fixes it with prompt structure and mid-conversation re-anchoring, backed by runnable code and measured results.
I Was Starting the Ad SDK Before the Consent Dialog — Fixing ATT and AdMob Ordering with Claude Code
Starting the ad SDK at launch initializes it before ATT consent, so the IDFA is all zeros and only your measurement goes thin while fill rate stays fine. Here is how I traced the scattered init paths with Claude Code and reshaped them to start ads only after ATT resolves.
Rolling Out Trusted Devices for a Small Team: Enrollment, Preflight, and Rotation
How to introduce Team/Enterprise Trusted Devices for a 2-5 person team: device enrollment, an unattended-run preflight gate, and closing the gaps that appear during device rotation and offboarding.
Fable 5 Is Back Worldwide and Sonnet 5 Is the Default — Where Each of the Three Models Belongs in a Solo Automation Stack
With Fable 5 redeployed worldwide and Sonnet 5 now the default, solo automation suddenly has three capable top-tier models to reach for. Instead of ranking them, this piece assigns each a role and captures that in a policy object with a fallback ladder and run-level logging.
When to Use Claude Code's Native 1M Context — and When Not To: A Cost-Based Rule
With Sonnet 5 as the default, Claude Code now handles a native 1M-token context. A big window is convenient, but every token you park in it is billed again each turn. Should you load the whole repo, or feed slices? Here is an estimable token model and a decision rule that gives a concrete answer per situation, with working code and the traps to avoid.