Articles for people who
ship the stack
Original rewrites on React, Node.js, TypeScript and AI — practical notes from the same engineering practice behind our operator software. Article bodies are in English.
Tagged: agent-harness
A Production Harness for Claude Code: Six Layers From Slack to Merged MR
How a six-layer harness turns Slack tickets into merged code: event intake, gated agent loops, durable state, self-healing, observability and operator control.
42 min readWhy the Agent Harness, Not the Model, Decides What Becomes Real
A timed-out CRM update shows why agents need tool contracts, unknown-outcome states, idempotent retries and verified completion, plus controls worth keeping.
6 min readSurviving the Fold: A Ladder for Long-Running Agent Sessions and Compaction
How long agent sessions stay affordable and coherent: the cache tail as rent, five rungs before compaction, delegation as a memory boundary, resume costs.
16 min readAgent Harnesses: A Folder Standard for Skills and References
How a harness extends agent skills with a HARNESS.md file, skill and reference directories, progressive disclosure, and nested layouts for large agents.
8 min readgrep Beat Vector Search in Agent Memory Tests, and the Harness Mattered as Much
A study of 116 LongMemEval questions found grep ahead of vector retrieval, but how a harness delivers results swung accuracy as much as the retriever. What to take from it.
7 min readDeepSeek Harness (dsh): The Open Agent Runtime That Rewrites Its Own Config
What DeepSeek Harness is, why its self-editing agent stands out, how the community reacted, and what to prepare for before running it with local models.
6 min readThe Five Layers of an Agent Harness, With Code That Fails Without Them
The five layers of an agent harness, each with code that fails without it: execution boundary, sandboxing, memory, verification loops and the context pipeline.
14 min readYour Agent Harness Is Probably Overengineered: What to Strip Out
Your agent harness is probably overengineered. As models improve, scaffolding turns into dead weight. How review bloat and context cost add up, and what to cut.
9 min readDesigning a RAG Pipeline So AI Agents Reason, Not Just Query
A RAG pipeline for AI agents should expose confidence, source diversity and coverage, so the agent can notice gaps, query again and explain how it got there.
5 min read
About these articles
Request a 24h estimate
Need the same stack in a production operator layer? Send the brief — estimate within 24 hours.