| name | improve-codebase-architecture |
| description | Surface architectural friction across the entire project and propose deepening opportunities — refactors that turn shallow modules into deep ones. |
| disable-model-invocation | true |
Improve Codebase Architecture
Surface architectural friction and propose deepening opportunities — refactors that turn shallow modules into deep ones. The aim is testability and AI-navigability.
Use this shared design vocabulary — stay on these terms, don't drift into
"component," "service," "API," or "boundary":
- module — a unit of code with a single responsibility
- interface — what a module exposes; the surface callers depend on
- depth — how much complexity a module hides behind its interface;
a deep module has a small interface hiding a large implementation
- seam — the place where two modules meet; their shared contract
- adapter — a module whose sole job is to translate between two seams
- leverage — how much work a module does relative to what a caller
must understand to use it
- locality — how close related code lives; code that changes together
should live together
Principles:
- Deletion test — would deleting this module concentrate complexity
(good, it's deep), or just move it (bad, it's shallow)?
- The interface is the test surface — if a module is hard to test,
its interface is wrong
- One adapter = hypothetical seam, two = real — the first adapter on a
seam is scaffolding; the second proves the seam is right
The domain language in docs/architecture.md and docs/data-model.md
names good seams. Create docs/glossary.md lazily when shared terminology
is needed across docs.
Process
1. Explore
Read the project's architecture docs (docs/architecture.md,
docs/data-model.md) first. If a CONTEXT.md domain glossary exists,
read that too.
Then use the Agent tool with subagent_type=Explore to walk the entire
codebase — not just recently changed files or the current branch diff.
Don't follow rigid heuristics — explore organically and note where you
experience friction:
- Where does understanding one concept require bouncing between many small modules?
- Where are modules shallow — interface nearly as complex as the implementation?
- Where have pure functions been extracted just for testability, but the real bugs hide in how they're called (no locality)?
- Where do tightly-coupled modules leak across their seams?
- Which parts of the codebase are untested, or hard to test through their current interface?
Apply the deletion test to anything you suspect is shallow: would deleting it concentrate complexity, or just move it? A "yes, concentrates" is the signal you want.
2. Present candidates
For each candidate found during exploration, present a structured summary:
- Files — which files/modules are involved
- Problem — why the current architecture is causing friction
- Solution — plain English description of what would change
- Benefits — explained in terms of locality and leverage, and how tests would improve
- Recommendation strength — one of
Strong, Worth exploring, Speculative
End with a Top recommendation section: which candidate you'd tackle first and why.
Use the domain vocabulary from docs/architecture.md,
docs/data-model.md, and the terms defined above. If docs/glossary.md
exists, use its terms — don't drift into "component," "service," "API," or
"boundary."
Existing decision conflicts: if a candidate contradicts a decision recorded in an existing doc, only surface it when the friction is
real enough to warrant revisiting that decision. Mark it clearly. Don't list every theoretical refactor a past decision forbids.
Do NOT propose interfaces yet. After presenting the candidates, ask the user: "Which of these would you like to explore?"
3. Deepen the chosen candidate
Once the user picks a candidate, use add-feature to record durable decisions as you work through the design with them:
- Naming a deepened module after a concept not yet documented? Add the
term to the relevant doc under
docs/. Create docs/glossary.md lazily
if a cross-cutting glossary is needed.
- Sharpening a fuzzy term during the conversation? Update the relevant
doc right there.
- User rejects the candidate with a load-bearing reason? Offer to record
it as a durable decision in the relevant doc, framed as: "Want me to
record this so future architecture reviews don't re-suggest it?" Only
offer when the reason would actually be needed by a future explorer to
avoid re-suggesting the same thing — skip ephemeral reasons ("not worth it
right now") and self-evident ones.
- Want to explore alternative interfaces for the deepened module? Design
two competing interfaces in parallel, compare them against the deletion
test and leverage, and record the winner.
Completion Criterion
The architecture review is complete when:
- the exploration covers the full codebase (not just recent changes),
- the candidates and their trade-offs are clearly communicated,
- the top recommendation is explicit,
- and any durable architecture decisions that emerge are recorded in the
relevant docs.