design-thinkinglisted
Install: claude install-skill widnyana/eyay-toolkits
# Design Thinking
The code method. Read the problem. Draw the data flow as a call graph. Write
code that IS the graph. If the code doesn't match the graph, something is wrong.
One of three methods with the same shape (see [design-graph.md](../../references/design-graph.md)
for the interface, [graph-protocol.md](../../references/graph-protocol.md) for delegation to
agents). The original is written in Effect-TS ([effect-ts.md](../../references/effect-ts.md)).
This version is stack-agnostic: where the original says `Effect`, `gen`, or
`pipe`, use your stack's equivalent (Result types, exceptions at layer joins,
middleware, RAII, constructor injection). The discipline never changes, only
the vocabulary.
**Hard gate — applies in every harness (pi, omp, Claude Code, any agent):**
When the user asks for a feature, change, fix, or refactor — or when any turn
will end in proposed code or file edits — you MUST render the graph (in the
[output format](../../references/output-format.md)) and any clarifying questions BEFORE
touching any file, and implement only after the user approves the design.
Reading and exploring code first is fine; the FIRST edit is not allowed
before the graph. This rule is not optional and not stack-dependent.
```
X → Graph → Effect<A, E, R>
│ │ │ │ │
│ │ │ │ └─ what each node needs (§5)
│ │ │ └──── where the graph breaks (§4)
│ │ └─────── what flows through nodes (§2)
│ │
│