Skip to content
Back to skills

Planning Codebase Simplification

ASecurity

Planning a whole-codebase simplification: audits a Go/TS codebase for needless abstraction and emits a KILL/REVIEW/KEEP plan plus a ring:running-dev-cycle task array. Plans only — no edits. Detects single-impl interfaces, pass-through shims, translation-free adapters, and dead-code cascade chains under an inverted burden of proof. Use for pre-public or post-pivot cleanup. Skip for current diff review (use ring:reviewing-code).

  • 215 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added September 6, 2026
developmentrustgocode-reviewapibackend

Works with

  • claude code
  • terminal
  • cli
  • api

Security analysis

A100/100

Scanned September 6, 2026

npx -y skills add LerianStudio/ring --skill planning-codebase-simplification --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Planning Codebase Simplification?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Planning Codebase Simplification
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/lerianstudio-planning-codebase-simplification/badge)](https://www.skillsdirectory.com/skills/lerianstudio-planning-codebase-simplification)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: ring:planning-codebase-simplification
description: "Planning a whole-codebase simplification: audits a Go/TS codebase for needless abstraction and emits a KILL/REVIEW/KEEP plan plus a ring:running-dev-cycle task array. Plans only — no edits. Detects single-impl interfaces, pass-through shims, translation-free adapters, and dead-code cascade chains under an inverted burden of proof. Use for pre-public or post-pivot cleanup. Skip for current diff review (use ring:reviewing-code)."
---

# Dev Simplify — Whole-Codebase Structural Sweep

## When to use
- User asks to simplify, flatten, or audit architecture of a whole codebase
- User mentions "too much indirection", "kill shims", "unnecessary abstractions"
- Pre-public application where break-compatibility refactor is cheap
- Post-pivot cleanup: speculative scaffolding accumulated during exploration

## Skip when
- Diff review on a feature branch → use ring:reviewing-code
- Standards-conformance refactor → use ring:planning-backend-refactor
- Dead code from a specific change → use ring:dead-code-reviewer in ring:reviewing-code
- Application already has external clients depending on internals

## Related
**Complementary:** ring:reviewing-code, ring:codebase-explorer
**Similar:** ring:planning-backend-refactor, ring:auditing-production-readiness


**Core principle:** DELETE is the default verdict. An abstraction survives only with concrete evidence of the swap it enables.

## Hard Constraint

Default: **public APIs MUST NOT break** (HTTP routes, SDK surface, webhooks, event contracts).

Supply `hard_constraint` input to override. Must be declared — never auto-inferred.

## Dispatch Protocol

### ⛔ STOP-CHECK BEFORE DISPATCH

Before emitting any Task call, count the explorers you intend to launch in this turn.
- Count MUST equal 6 (or 5 if branch has no commits ahead of main — Task 5 skipped).
- If your dispatch count diverges → STOP and reconcile against the task table below.
- No substitutions, no omissions.

### ⛔ MUST NOT trickle-dispatch

All explorers leave in the SAME TURN, before reading any explorer output.

Forbidden sequences:
- Dispatch explorer 1 → read result → dispatch explorer 2
- Dispatch a subset → wait → dispatch the rest
- Dispatch follow-up explorers conditioned on partial output
- Loop sequentially over the task list

If you find yourself about to dispatch an explorer in a turn AFTER any explorer has already returned a result → STOP. You violated parallel dispatch. Report the violation and mark the phase INCOMPLETE rather than completing the trickle.

### Self-verify after dispatch

After the dispatch turn, verify all scoped Task calls (6, or 5 if Task 5 skipped) were emitted in that single turn. If fewer went out than scoped, the phase did NOT execute correctly. Mark INCOMPLETE and surface the dispatch failure — do NOT silently continue with a partial pool.

### Parallel dispatch — atomic batch

Emit all scoped Task calls (the count established in the STOP-CHECK above — 6 or 5) in a SINGLE TURN, as one atomic batch.

**If your runtime exposes a `multi_tool_use.parallel` wrapper**, use it to dispatch the complete pool in one wrapped invocation. This is the canonical fan-out mechanism on OpenAI-style tool envelopes and on certain Anthropic SDK consumers — naming it explicitly activates parallel emission on runtimes where trickle-dispatch is the default behavior.

**If your runtime emits parallel tool_use blocks natively** (Claude Code with Claude models), `multi_tool_use.parallel` may not be needed — but naming it is harmless and serves as an enforcement anchor.

The STOP-CHECK, anti-trickle, and self-verify guards above remain binding regardless of which mechanism your runtime uses.

Dispatch 6 explorer agents in **parallel** (5 if branch has no commits ahead of main — skip Task 5):

| Task | Agent | Focus |
|------|-------|-------|
| 1a | ring:codebase-explorer | Single-impl interfaces, ports, repositories |
| 1b | ring:codebase-explorer | Speculative factories, builders, strategies, facades |
| 2 | ring:codebase-explorer | Translation-free adapters, pass-through shims, internal DTOs |
| 3 | ring:codebase-explorer | Architecture topology mapping, indirection depth |
| 4 | ring:codebase-explorer | Cascade chains (Three Rings applied to codebase) |
| 5 | ring:codebase-explorer | Branch AI slop (diff vs main) — skip if no commits ahead |

**Explorer dispatch contract:**

```
## Target: <absolute path to repo root>
## Your Focus: <smell category from task table above>
## Hard Constraint: {hard_constraint}
## Output: Write to /tmp/simplify-{task}-findings.json
Schema: { task, findings: [{name, file_line, smell, rebuttal_if_any, blast_radius, public_api_impact, action}], cascade_chains: [...] }
```

## Abstraction Smell Rubric

| Smell | Signal | Default Action |
|---|---|---|
| Single-implementation interface | One concrete impl; test doubles identical to prod | DELETE |
| Translation-free adapter | A→B is rename-only, 1:1 field mapping | DELETE |
| Pass-through shim | Wraps call site-for-site, no cross-cutting concern | DELETE |
| Speculative factory/builder | Always constructs the same concrete type | DELETE |
| One-strategy strategy | Dispatch over enum with one case | COLLAPSE |
| One-consumer facade | Single call site, "for future reuse" | COLLAPSE INTO CALLER |
| Config seam over constant | Indirection for a value that never varies | DELETE |
| Internal DTO ↔ entity with 1:1 fields | Translation across identical shapes | DELETE |
| Hexagonal port with one adapter | No swap pressure | COLLAPSE |
| Narrating comment (branch diff) | Comment restates what code literally does | DELETE |
| Defensive check in trusted path (branch diff) | Guard where caller already validated | DELETE |

**Accepted evidence to KEEP** (must name concretely, not hypothetically):
- Second implementation exists today in this repo
- Swappability exercised in tests with divergent behavior
- Cross-process or cross-language boundary
- Regulatory or contractual requirement

## Phase 4: Consolidated Report

Dispatch synthesizer to read all explorer files and emit:
1. `docs/ring:planning-codebase-simplification/simplify-report-{timestamp}.md` — KILL / REVIEW / KEEP tables
2. `docs/ring:planning-codebase-simplification/simplify-tasks-{timestamp}.json` — ring:running-dev-cycle task array

MUST emit both artifacts. MUST include cascade chains decomposed into per-ring tasks with `depends_on` wiring.

## Output Format

```markdown
## Simplify Summary
- Scope / Hard Constraint / Generated
- Kill list: N items | Review list: N items | Keep list: N items

## Hard Constraint
- Declared constraint + load-bearing surface location

## Kill List
| Name | file:line | Smell | Blast radius | Action | Acceptance Criteria |

## Review List
| Name | file:line | Smell | Why uncertain | Recommended next step |

## Keep List
| Name | file:line | Smell resembled | Evidence |

## Cascade Chains
| Chain ID | Leaf | Ring depth | Terminal type | Collapse blast radius |

## Cascade Execution Plan
Per-chain DAG: ring-1 (leaf) → ring-N, each with depends_on wiring

## Remaining Risks
| Risk ID | Related findings | Risk type | Mitigation |
```

## Task JSON Schema

```json
{
  "tasks": [{
    "id": "simplify-001",
    "title": "",
    "severity": "KILL | REVIEW",
    "smell_category": "unexercised-seam | speculative-construction | translation-layer | topology | cascade | branch-slop",
    "files_affected": [],
    "blast_radius": {"files": 0, "lines": 0},
    "acceptance_criteria": [],
    "estimated_complexity": "trivial | moderate | complex",
    "depends_on": [],
    "rebuttal_if_kept": null
  }]
}
```

Cascade chains → N tasks with `depends_on` wiring (leaf = ring-1, `depends_on: []`).

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.