This article is published in English.
Practical notes: I Don’t Trust AI Coding Agents by Default — Here’s the 8-Layer
Operable walkthrough of Practical notes: I Don’t Trust AI Coding Agents by Default — Here’s the 8-Layer: contracts, checks, and drop-in code slots for teams shipping this pattern.
The following notes reconstruct a practical path around “I Don’t Trust AI Coding Agents by Default — Here’s the 8-Layer Claude Code Setup That Earns It ”. Emphasis stays on contracts, checks, and drop-in code placeholders rather than motivational framing. When working through the Overview stage, write down the contract first: required inputs, success signal, and what happens on partial failure. That checklist keeps later code changes honest. Document the happy path and the recovery path together. Retries, human gates, and dead-letter handling are part of the product, not later polish.
Layer 1: The CLAUDE.md Hierarchy — Global Rules vs. Project-Specific Instructions
The Layer 1 The CLAUDE stage works best when treated as a measurable surface. Capture one golden transcript, one failure case, and the rollback note before expanding scope. Prefer small, testable units over sprawling scripts. When a step fails, the failure should point at a single responsibility rather than a tangled pipeline. Keep graph state flat and typed. Nested blobs hide which node wrote which field and break resume after interrupts.
Layer 2: Claude Code’s Auto Memory — Persistent Context Beyond CLAUDE.md
The Layer 2 Claude Code stage works best when treated as a measurable surface. Capture one golden transcript, one failure case, and the rollback note before expanding scope. Treat this stage as a contract between inputs and validated outputs. Name the artifacts, define success checks, and refuse silent partial completion. Keep graph state flat and typed. Nested blobs hide which node wrote which field and break resume after interrupts.
Layer 3: Guardrail Hooks — Deterministic Control Over an AI Coding Agent
The Layer 3 Guardrail Hooks stage works best when treated as a measurable surface. Capture one golden transcript, one failure case, and the rollback note before expanding scope. Record timings and token or query cost next to functional results. Cost visibility early prevents surprise bills when the path moves from demo to shared environments. Keep graph state flat and typed. Nested blobs hide which node wrote which field and break resume after interrupts. The Layer 3 Guardrail Hooks stage works best when treated as a measurable surface. Capture one golden transcript, one failure case, and the rollback note before expanding scope. Document the happy path and the recovery path together. Retries, human gates, and dead-letter handling are part of the product, not later polish.
Layer 4: Code Intelligence Over Grep-and-Read Loops
For the Layer 4 Code Intelligence stage, define the inputs, the owner of the step, and the exit criteria before changing code. Operators should be able to re-run the step from a known checkpoint without guessing hidden state. Prefer small, testable units over sprawling scripts. When a step fails, the failure should point at a single responsibility rather than a tangled pipeline. Put human approval on edges that spend money or change production data. Compile-time wiring does not equal business completeness.
Layer 5: Live Device Verification for Flutter and Mobile Development
For the Layer 5 Live Device stage, define the inputs, the owner of the step, and the exit criteria before changing code. Operators should be able to re-run the step from a known checkpoint without guessing hidden state. Treat this stage as a contract between inputs and validated outputs. Name the artifacts, define success checks, and refuse silent partial completion. Put human approval on edges that spend money or change production data. Compile-time wiring does not equal business completeness.
Layer 6: Deferred Tool Loading and MCP Context Management
For the Layer 6 Deferred Tool stage, define the inputs, the owner of the step, and the exit criteria before changing code. Operators should be able to re-run the step from a known checkpoint without guessing hidden state. Record timings and token or query cost next to functional results. Cost visibility early prevents surprise bills when the path moves from demo to shared environments. Authenticate at the gateway and re-authorize at the data plane. A bearer token alone is not a tenancy boundary. For the Layer 6 Deferred Tool stage, define the inputs, the owner of the step, and the exit criteria before changing code. Operators should be able to re-run the step from a known checkpoint without guessing hidden state. Document the happy path and the recovery path together. Retries, human gates, and dead-letter handling are part of the product, not later polish.
Layer 7: Claude Code Skills and Subagents you Actually Use
When working through the Layer 7 Claude Code stage, write down the contract first: required inputs, success signal, and what happens on partial failure. That checklist keeps later code changes honest. Prefer small, testable units over sprawling scripts. When a step fails, the failure should point at a single responsibility rather than a tangled pipeline. Checkpoint after expensive steps. Resume should not re-bill the same LLM call when an operator retries a later node.
Layer 8: Where you Draw the Line on AI Agent Autonomy
When working through the Layer 8 Where you stage, write down the contract first: required inputs, success signal, and what happens on partial failure. That checklist keeps later code changes honest. Treat this stage as a contract between inputs and validated outputs. Name the artifacts, define success checks, and refuse silent partial completion. Checkpoint after expensive steps. Resume should not re-bill the same LLM call when an operator retries a later node.
Why a Real Claude Code Setup Actually Matters
When working through the Why a Real Claude stage, write down the contract first: required inputs, success signal, and what happens on partial failure. That checklist keeps later code changes honest. Record timings and token or query cost next to functional results. Cost visibility early prevents surprise bills when the path moves from demo to shared environments. Checkpoint after expensive steps. Resume should not re-bill the same LLM call when an operator retries a later node. When working through the Why a Real Claude stage, write down the contract first: required inputs, success signal, and what happens on partial failure. That checklist keeps later code changes honest. Document the happy path and the recovery path together. Retries, human gates, and dead-letter handling are part of the product, not later polish.
Operational checklist
For the Operational checklist stage, define the inputs, the owner of the step, and the exit criteria before changing code. Operators should be able to re-run the step from a known checkpoint without guessing hidden state.
Keep configuration outside application code. Environment files, secret stores, and feature flags belong in one place operators can audit without reading the whole graph.
Put human approval on edges that spend money or change production data. Compile-time wiring does not equal business completeness.
Write a short runbook: how to rotate keys, how to drain the queue, how to roll back the last ingest.
Document the happy path and the recovery path together. Retries, human gates, and dead-letter handling are part of the product, not later polish.
Put human approval on edges that spend money or change production data. Compile-time wiring does not equal business completeness.
Before promoting the stack, freeze versions, capture a golden transcript for the critical path, and confirm rollback steps. Shared environments need rate limits, tenancy checks, and a clear owner for secret rotation. Prefer boring reliability over clever one-off demos.
Batch note for c85223e176f2: keep provider keys out of the repo, set a per-session token ceiling, and store transcripts next to the eval fixtures so later model swaps stay comparable.
The hardening note 0 stage works best when treated as a measurable surface. Capture one golden transcript, one failure case, and the rollback note before expanding scope. Keep configuration outside application code. Environment files, secret stores, and feature flags belong in one place operators can audit without reading the whole graph.
Hardening detail 0/770: measure wall time, error class, and token spend for this note, then decide whether to keep the change based on a fixed question set rather than anecdote.
For the hardening note 1 stage, define the inputs, the owner of the step, and the exit criteria before changing code. Operators should be able to re-run the step from a known checkpoint without guessing hidden state. Prefer small, testable units over sprawling scripts. When a step fails, the failure should point at a single responsibility rather than a tangled pipeline.
Hardening detail 1/770: measure wall time, error class, and token spend for this note, then decide whether to keep the change based on a fixed question set rather than anecdote.
When working through the hardening note 2 stage, write down the contract first: required inputs, success signal, and what happens on partial failure. That checklist keeps later code changes honest. Record timings and token or query cost next to functional results. Cost visibility early prevents surprise bills when the path moves from demo to shared environments.
Hardening detail 2/770: measure wall time, error class, and token spend for this note, then decide whether to keep the change based on a fixed question set rather than anecdote.
The hardening note 3 stage works best when treated as a measurable surface. Capture one golden transcript, one failure case, and the rollback note before expanding scope. Document the happy path and the recovery path together. Retries, human gates, and dead-letter handling are part of the product, not later polish.
Hardening detail 3/770: measure wall time, error class, and token spend for this note, then decide whether to keep the change based on a fixed question set rather than anecdote.
For the hardening note 4 stage, define the inputs, the owner of the step, and the exit criteria before changing code. Operators should be able to re-run the step from a known checkpoint without guessing hidden state. Treat this stage as a contract between inputs and validated outputs. Name the artifacts, define success checks, and refuse silent partial completion.
Hardening detail 4/770: measure wall time, error class, and token spend for this note, then decide whether to keep the change based on a fixed question set rather than anecdote.
When working through the hardening note 5 stage, write down the contract first: required inputs, success signal, and what happens on partial failure. That checklist keeps later code changes honest. Keep configuration outside application code. Environment files, secret stores, and feature flags belong in one place operators can audit without reading the whole graph.
Hardening detail 5/770: measure wall time, error class, and token spend for this note, then decide whether to keep the change based on a fixed question set rather than anecdote.