The grid

The practices of this library come from all of software engineering: architecture, types, tests, refactoring, tooling. Birgitta Böckeler's grid, from Harness engineering for coding agent users (martinfowler.com, April 2026), situates each of them in work with a coding agent: where the knowledge acts, and who applies it.

Guides act before the agent: they "anticipate the agent's behaviour and aim to steer it before it acts". Sensors act after it: they "observe after the agent acts and help it self-correct".

Each is computational, "deterministic and fast, run by the CPU", or inferential, "semantic analysis, AI code review, 'LLM as judge'", slower, more expensive and less deterministic.

Inferential: a model judgesComputational: a program decides
Guides, before the agent actsInferential guideComputational guide
Sensors, after the agent actsInferential sensorComputational sensor

Each cell's page lists examples of what acts there, Böckeler's and the ones this library adds, then the practices of this library that act there when an agent writes the code.

+ Propose a practice