improve-codebase-architecture
Find codebase deepening opportunities, present a visual HTML review, then guide selected design.
Skill by davidbasilefilho
Chalk Labs
Need help deploying improve-codebase-architecture?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/davidbasilefilho-improve-codebase-architecture
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: design, frontend
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/improve-codebase-architecture
What's inside
- Skill name
improve-codebase-architecture- Description
- Find codebase deepening opportunities, present a visual HTML review, then guide selected design. Use for architectural friction, shallow modules, weak seams, poor locality, or testability-driven refactoring.
- License
- Apache-2.0
- Instructions
- 44 lines in SKILL.md · view on GitHub
Files (2)
- HTML-REPORT.md
- SKILL.md
SKILL.md (first lines)
# Improve Codebase Architecture Seek deep modules: small interfaces hiding substantial implementation, improving testability and AI navigation. Use `codebase-design` vocabulary exactly: **module**, **interface**, **depth**, **seam**, **adapter**, **leverage**, **locality**. Apply deletion test and principles: interface is test surface; one adapter is hypothetical seam, two are real. Use `CONTEXT.md` domain terms and respect `docs/adr/`. ## 1. Explore Read domain glossary and relevant ADRs. Explore organically for: - understanding scattered across small modules; - interfaces nearly as complex as implementations; - pure functions extracted for tests while call-site bugs remain; - coupled modules leaking across seams; - untested or hard-to-test interfaces. Deletion test: would removing candidate concentrate complexity rather than move it? Only then treat as strong deepening opportunity. ## 2. Report Create self-contained `<temp>/architecture-review-<timestamp>.html`; never write report into repo. Resolve `$TMPDIR`, `/tmp`, or `%TEMP%`. Open with platform command and provide absolute path. Use Tailwind CDN, Mermaid CDN for graph-shaped relationships, and CSS/SVG for editorial visuals. Follow [HTML-REPORT.md](HTML-REPORT.md). Every candidate card needs: - files/modules; - friction; - proposed change in plain language; - locality, leverage, and test benefits; - side-by-side before/after visual; - `Strong`, `Worth exploring`, or `Speculative` badge. End with top recommendation and reason. Use domain vocabulary. Mark worthy ADR conflicts explicitly; omit theoretical conflicts. Do not design interfaces yet. Ask which candidate user wants to explore. ## 3. Design Selected Candidate Run `grilling` for constraints, dependencies, deepened module shape, hidden implementation, seam, and surviving tests. Keep artifacts current as decisions form: - New/sharpened domain concept: update `CONTEXT.md`; create lazily. - Rejection with durable, load-bearing reason: offer ADR; skip temporary/self-evident reasons. - Alternative interfaces: run `codebase-design` and its design-it-twice parallel pattern.
From davidbasilefilho/codex-skills, Apache-2.0.
More from davidbasilefilho/codex-skills
- 0.468score
diagnosing-bugs
davidbasilefilho
Diagnose hard bugs and performance regressions through a strict reproduce-minimise-hypothesise-instrument-fix loop.
SkillCFreshness: CClaude Code +9 - 0.466score
tdd
davidbasilefilho
Apply test-driven development for features or fixes.
SkillCFreshness: CClaude Code +9 - 0.466score
compress
davidbasilefilho
Rewrite any text, skill, prompt, documentation, instructions, prose, notes, or structured content to use fewer words and tokens while preserving meaning…
SkillCFreshness: CClaude Code +9 - 0.466score
high-end-visual-design
davidbasilefilho
Design premium agency-level web interfaces with deliberate typography, spacing, surfaces, responsive layouts, micro-interactions, and performant motion.
SkillCFreshness: CClaude Code +9 - 0.466score
controlled-parallel-work
davidbasilefilho
Complete software work with the explorer and forge custom agents under tight scope.
SkillCFreshness: CClaude Code +9 - 0.466score
define-goal
davidbasilefilho
Define or refine a concrete, measurable goal before work starts.
SkillCFreshness: CClaude Code +9 - 0.463score
brandkit
davidbasilefilho
Generate premium brand-kit images, guideline boards, logo systems, identity decks, and visual-world presentations.
SkillCFreshness: CClaude Code +9 - 0.463score
context7
davidbasilefilho
Retrieve current library, framework, and component documentation through Context7.
SkillCFreshness: CClaude Code +9 - 0.463score
gpt-taste
davidbasilefilho
Create high-variance editorial UI with AIDA structure, wide heroes, gapless bento grids, advanced GSAP motion, varied assets, and strict preflight verification.
SkillCFreshness: CClaude Code +9 - 0.463score
review-plan
davidbasilefilho
Review an implementation plan against repository evidence, fix concrete planning failures, then stop before implementation.
SkillCFreshness: CClaude Code +9 - 0.463score
find-skills
davidbasilefilho
Discover and install agent skills when users ask how to do something, request a skill, seek specialized capability, or want to extend agent functionality.
SkillCFreshness: CClaude Code +9 - 0.463score
teach
davidbasilefilho
Teach a skill or concept through a persistent workspace of mission-driven lessons, references, resources, practice, and learning records.
SkillCFreshness: CClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Jul 12, 2026 (3 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- Apache-2.0
- Repository
- github.com/davidbasilefilho/codex-skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from davidbasilefilho
- 0.504score
security-research
davidbasilefilho
Exploitability-driven security audit: surface, auth/data, runtime/supply, and independent PoCs; not routine review.
SkillBFreshness: BClaude Code +9 - 0.504score
frontend-taste
davidbasilefilho
Design/audit distinctive production UI using the complete pols.dev anti-slop law; user visual direction wins.
SkillBFreshness: BClaude Code +9 - 0.499score
debugging
davidbasilefilho
Diagnose unexplained reproducible failures via competing hypotheses and focused experiments.
SkillBFreshness: BClaude Code +9 - 0.499score
caveman
davidbasilefilho
Simplest direct implementation satisfying proven requirements; resists speculative architecture.
SkillBFreshness: BClaude Code +9 - 0.499score
context7-cli
davidbasilefilho
Context7 lookup for version-sensitive library APIs; use librarian for broader research.
SkillBFreshness: BClaude Code +9 - 0.499score
babysit-ci
davidbasilefilho
Watch a specified CI run to completion; diagnose and, when authorized, fix causal failures.
SkillBFreshness: BClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get improve-codebase-architecture running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
improve-codebase-architecture skill. Agent Mart. https://agentmart.dev/skills/davidbasilefilho-improve-codebase-architecture (last updated 2026-10-24)