Write AGENTS.md From Your Transcripts, Not Imagined Failures
Raw frequency tells you which failures are real. It still ranks them wrong. Weight each promoted category by what one instance costs, and the rule that looked #2 by count drops to last.
Engineering Blog
Technical insights into multi-agent orchestration, local protocols, and
developer-centric terminal workflows.
Hooks That Validate AI Agent Output Before It Ships
Against a 1-in-12 hallucination rate on a checkable failure, a deterministic hook with one capped retry drives the shipped-wrong rate to exactly zero and moves the remaining risk onto one computable number: the halt tax.
MCP Resources, Tools, and Prompts: the Three Primitives
The MCP spec hands a server three levers: resources for the application, tools for the model, prompts for the user. Skip that sort and your reads ride the tool list every turn, at a price you can compute.
MCP Elicitation: Have Tools Ask for Missing Parameters
When a required tool parameter is a fact only the human holds, the two familiar responses, fail and guess, each cost one round trip on the wire and guarantee nothing. Under the MCP spec current since 2026-07-28, asking costs exactly two round trips, fixed, whatever the human answers.
Git Pre-Commit Hooks Stop AI Agents from Committing Bugs
A soft AGENTS.md rule is a per-commit probability, and probabilities compound: 95% compliance becomes a 92.31% chance of a lapse somewhere in a 50-commit sprint. The hook most teams write only cuts that to 22.37%. Here is the arithmetic, layer by layer, and the one layer that zeroes it.
Have an MCP Server Grade the Git Diff, Not Your Summary
Ask an agent to grade the work you just described and it grades the paragraph, not the code, because agreeing with a confident claim is the model's strongest reflex. Hand the same model the live diff and a rubric you author, and it finds the four misses your summary hid.
Reproduce AI Agent Bugs as Synthetic Test Scenarios
A one-line bug report about an agent carries exactly one testable fact: how often the failure fires. Turn the report into a committed synthetic scenario, then let that fact size how many reruns a fixed verdict is worth.
RAG Debugging: Retrieval Failures vs Generation Failures
A confidently wrong answer is two bugs in one costume: a fact that never reached the context window, and a fact that arrived and got ignored. Log the window, score two binaries, and each bug routes to a different layer of the stack.
Stop Broken Function Level Authorization in AI-Written Code
Every inline authorization check an agent writes is an independent guess at the same policy. The guesses disagree, quietly. Here is the arithmetic of how fast that compounds, and the one call that stops it.
Slash Commands and Rules: Stop Retyping the Same AI Correction
The correction you keep re-typing into the chat is a workflow you haven't written down. Priced with a small invented schedule, it costs about 43 silently skipped build gates a quarter. Encoded as a slash command, it costs zero.