The Skill Call That Returned Nothing — Pinning Down Name Resolution Before an Unattended Run
A skill whose name collides with a built-in terminal command may never resolve in a non-interactive session. Here is a preflight that builds the effective-name table before launch, plus the measured run where the checker itself silently skipped two entries.
The MCP Server Connects Fine and Still Hands Back Zero Tools — Catching Silent Capability Drift with a Tool Manifest Diff
A missing credential does not break the MCP handshake. initialize succeeds, tools/list succeeds, and the array comes back empty. Here is the measured behaviour, and a preflight that locks the expected tool surface and fails closed before the agent ever starts.
I Believed Plan Mode Only Read — Replacing That Belief With Machinery
Claude Code 2.1.212 fixed a bug where plan mode ran file-modifying Bash commands without the permission prompt or the SDK canUseTool callback. Here is what could happen while that assumption was broken, how to verify your own setup, and how to stop leaning on a mode name for safety.
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.
A 1M Context Window Is the New Default — So I Built an Admission Policy Instead of Filling It
Sonnet 5 is now the Claude Code default and native 1M context is standard. The hard errors disappeared, but a quieter kind of degradation took their place. Here is how I made it visible with a probe, plus an admission policy and an effective-token-cost view — with working code and my own measurements.
Which Model Ran Last Night's Unattended Session? Building Model Attribution and Default-Drift Detection After the Sonnet 5 Switch
Claude Code's default model switched to Sonnet 5, and unpinned headless runs changed models silently. Here is a working design for extracting the actual model from run output, appending an atomic run record, and deciding per task lineage whether to pin or follow the default.
Don't Accept an Agent's Numbers and Citations As-Is — A Verification Gate Built on a Dedicated Auditor Subagent
A design that verifies every number and citation in an agent-generated summary using a separate subagent before accepting it — with working TypeScript for deterministic recomputation and fail-closed source matching.
When You Fan Out Streaming Sessions, Your Laptop's CPU Gives Out First — An Adaptive Throttle That Caps Concurrency by Measured Load
Even with lighter streaming, fanning out many sessions on one machine saturates the host CPU before anything else. Here is why a fixed semaphore fails, plus a working adaptive gate that raises and lowers concurrency from measured CPU.