
Gsd
Spawn a focused mapper agent that explores your repo and writes prescriptive analysis files under `.planning/codebase/` before you plan or implement on brownfield code.
Overview
gsd (GSD Codebase Mapper) is an agent skill most often used in Build (also Validate scope prep and Ship review context) that explores a codebase by focus area and writes structured analysis markdown into `.planning/codeb
Install
npx skills add https://github.com/ctsstc/get-shit-done-skills --skill gsdWhat is this skill?
- Four spawn focus areas: tech, arch, quality, and concerns—each maps to specific markdown deliverables
- Writes directly to `.planning/codebase/` so the orchestrator keeps a lighter context window
- Tech focus produces STACK.md and INTEGRATIONS.md; arch produces ARCHITECTURE.md and STRUCTURE.md
- Quality focus documents CONVENTIONS.md and TESTING.md; concerns captures debt in CONCERNS.md
- Prescriptive HOW-it-works style after thorough Read, Grep, Glob, and Bash exploration
- 4 focus areas: tech, arch, quality, concerns
- 8 named output documents across focus areas
Adoption & trust: 2.1k installs on skills.sh; 7 GitHub stars; 1/3 security scanners passed (skills.sh audits); trending (+100% hot-view momentum).
What problem does it solve?
You are about to change an existing repo but lack a durable, agent-readable map of stack, architecture, conventions, and technical debt.
Who is it for?
Brownfield solo projects where you spawn map-codebase with tech, arch, quality, or concerns before a GSD phase or large agent task.
Skip if: Greenfield repos with almost no code, or when you only need a one-off file read instead of four-category structured maps.
When should I use this skill?
You need to understand an existing codebase, start brownfield work, document stack or architecture, identify technical debt, or prepare codebase context for a new phase—triggered via map-codebase / analyze codebase / exp
What do I get? / Deliverables
You get prescriptive analysis documents in `.planning/codebase/` for the chosen focus so planning and coding agents can proceed without re-scanning the entire tree.
- STACK.md and INTEGRATIONS.md (tech)
- ARCHITECTURE.md and STRUCTURE.md (arch)
- CONVENTIONS.md and TESTING.md (quality)
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Brownfield understanding is canonical on the Build shelf because solo builders need stack, architecture, and debt context before they extend the product. Outputs are structured documentation artifacts (STACK, ARCHITECTURE, CONVENTIONS, CONCERNS), not runtime code—so docs is the best primary subphase.
Where it fits
Run the concerns focus before committing to a rewrite so scope reflects real debt and integration risk.
Spawn arch focus to produce ARCHITECTURE.md and STRUCTURE.md before an agent implements a new module.
Use tech focus output as input for a GSD phase that depends on external integrations listed in INTEGRATIONS.md.
Refresh quality focus CONVENTIONS.md and TESTING.md so review agents judge PRs against how the repo actually works.
Re-run concerns focus after a release cycle to update CONCERNS.md for the next iteration plan.
How it compares
Use instead of pasting directory listings into chat—this is a procedural mapper with fixed output filenames, not generic summarization.
Common Questions / FAQ
Who is gsd for?
Solo and indie builders using agent orchestration (especially GSD-style planning) who need structured codebase intelligence before they commit to a build or ship phase.
When should I use gsd?
When you need to understand structure on an existing project, document stack and integrations, capture conventions and testing patterns, or list concerns before a new phase; also when starting brownfield work in Validate-style scoping or re-baselining during Operate iteration.
Is gsd safe to install?
It uses Read, Grep, Glob, Bash, and Write on your filesystem; review the Security Audits panel on this Prism page and restrict spawn permissions if you are uncomfortable with shell and write access.
SKILL.md
READMESKILL.md - Gsd
# GSD Codebase Mapper Explores a codebase for a specific focus area and writes analysis documents directly to `.planning/codebase/`. ## When to Use Use this agent when: - You need to understand an existing codebase structure - Starting work on a brownfield project (existing code) - Need to document technology stack, architecture, or patterns - Want to identify technical debt and concerns - Preparing for a new phase that requires codebase context ## Focus Areas The mapper is spawned with one of four focus areas: - **tech** - Analyze technology stack and external integrations → writes STACK.md and INTEGRATIONS.md - **arch** - Analyze architecture and file structure → writes ARCHITECTURE.md and STRUCTURE.md - **quality** - Analyze coding conventions and testing patterns → writes CONVENTIONS.md and TESTING.md - **concerns** - Identify technical debt and issues → writes CONCERNS.md ## Core Responsibilities 1. **Explore thoroughly** - Read relevant files, understand patterns, identify key components 2. **Write directly** - Create documents in `.planning/codebase/` to reduce orchestrator context load 3. **Be prescriptive** - Document HOW things are done, not just WHAT exists 4. **Include file paths** - Every finding should have a file path in backticks for navigation 5. **Return confirmation only** - Don't return document contents, just confirm what was written ## Why This Matters These documents are consumed by other GSD commands: **`/gsd:plan-phase`** loads relevant codebase docs when creating implementation plans: | Phase Type | Documents Loaded | |------------|------------------| | UI, frontend, components | CONVENTIONS.md, STRUCTURE.md | | API, backend, endpoints | ARCHITECTURE.md, CONVENTIONS.md | | database, schema, models | ARCHITECTURE.md, STACK.md | | testing, tests | TESTING.md, CONVENTIONS.md | | integration, external API | INTEGRATIONS.md, STACK.md | | refactor, cleanup | CONCERNS.md, ARCHITECTURE.md | | setup, config | STACK.md, STRUCTURE.md | **`/gsd:execute-phase`** references codebase docs to: - Follow existing conventions when writing code - Know where to place new files (STRUCTURE.md) - Match testing patterns (TESTING.md) - Avoid introducing more technical debt (CONCERNS.md) ## Process ### Step 1: Parse Focus Area Read the focus area from your prompt. It will be one of: `tech`, `arch`, `quality`, `concerns`. ### Step 2: Determine Output Documents Based on focus, determine which documents to write: - `tech` → STACK.md, INTEGRATIONS.md - `arch` → ARCHITECTURE.md, STRUCTURE.md - `quality` → CONVENTIONS.md, TESTING.md - `concerns` → CONCERNS.md ### Step 3: Explore Codebase Use appropriate exploration commands for your focus area: **For tech focus:** ```bash # Package manifests ls package.json requirements.txt Cargo.toml go.mod pyproject.toml 2>/dev/null cat package.json 2>/dev/null | head -100 # Config files ls -la *.config.* .env* tsconfig.json .nvmrc .python-version 2>/dev/null # Find SDK/API imports grep -r "import.*stripe\|import.*supabase\|import.*aws\|import.*@" src/ --include="*.ts" --include="*.tsx" 2>/dev/null | head -50 ``` **For arch focus:** ```bash # Directory structure find . -type d -not -path '*/node_modules/*' -not -path '*/.git/*' | head -50 # Entry points ls src/index.* src/main.* src/app.* src/server.* app/page.* 2>/dev/null # Import patterns to understand layers grep -r "^import" src/ --include="*.ts" --include="*.tsx" 2>/dev/null | head -100 ``` **For quality focus:** ```bash # Linting/formatting config ls .eslintrc* .prettierrc* eslint.config.* biome.json 2>/dev/null cat .prettierrc 2>/dev/null # Test files and config ls jest.config.* vitest.conf