architecture-decision-records
Document technical decisions as Architecture Decision Records (ADRs) with context, options considered, and rationale.
Install
npx skills add https://github.com/spencerpauly/awesome-cursor-skills/tree/main/resources/architecture-decision-records
claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install spencerpauly-awesome-cursor-skills@llmmart
git clone https://github.com/spencerpauly/awesome-cursor-skills.git
The skills CLI installs just this skill, for any of its supported agents. Claude Code installs the whole spencerpauly/awesome-cursor-skills collection as a plugin from our marketplace. Git is the plain clone.
Skill manifest
Architecture Decision Records
Document significant technical decisions so future you (and your team) understands why choices were made.
When to Write an ADR
Write one when you're making a decision that:
- Is hard to reverse later
- Affects multiple parts of the system
- Involves tradeoffs between valid options
- Will be questioned by someone in 6 months
Examples: choosing a database, adopting a framework, changing auth strategy, restructuring the API, adding a build tool.
Template
Create files in docs/decisions/ (or adr/) with sequential numbering:
# ADR-001: Use PostgreSQL for primary database
## Status
Accepted | Proposed | Deprecated | Superseded by ADR-XXX
## Date
2026-04-10
## Context
What is the problem or situation that requires a decision?
Include constraints, requirements, and forces at play.
## Options Considered
### Option A: PostgreSQL
- Pros: ACID compliance, JSON support, mature ecosystem, free
- Cons: Requires managing connections, vertical scaling limits
### Option B: MongoDB
- Pros: Flexible schema, horizontal scaling
- Cons: No transactions across collections, eventual consistency issues
### Option C: PlanetScale (MySQL)
- Pros: Serverless, branching, managed
- Cons: Vendor lock-in, no foreign keys in their branching model
## Decision
We chose **PostgreSQL** because:
1. Our data is relational — users, teams, projects with clear relationships
2. We need ACID transactions for billing operations
3. JSON columns give us schema flexibility where needed
4. Prisma/Drizzle have excellent Postgres support
## Consequences
- We need to manage connection pooling (use PgBouncer or Prisma's built-in pool)
- Migrations must be backwards-compatible for zero-downtime deploys
- We accept vertical scaling limits and will shard later if needed
Workflow
- Identify the decision being made
- Research the options (at least 2-3 alternatives)
- Write the ADR using the template
- Review with the team (PR or discussion)
- Merge as accepted
- Reference from code comments when relevant:
// See ADR-001
File Naming
docs/decisions/
├── 001-use-postgresql.md
├── 002-adopt-trpc-over-rest.md
├── 003-switch-to-pnpm.md
└── template.md
Tips
- Keep ADRs short — 1-2 pages max
- Write them in the present tense ("We choose X" not "We chose X")
- It's OK to write an ADR after the fact if you forgot to write one during the decision
- Superseded ADRs should link to their replacement, not be deleted
- ADRs are not design docs — they capture the decision, not the full design
Files (awesome-cursor-skills)
-
SKILL.md 2.8 KB
--- name: architecture-decision-records description: Document technical decisions as Architecture Decision Records (ADRs) with context, options considered, and rationale. user-invocable: true --- # Architecture Decision Records Document significant technical decisions so future you (and your team) understands why choices were made. ## When to Write an ADR Write one when you're making a decision that: - Is hard to reverse later - Affects multiple parts of the system - Involves tradeoffs between valid options - Will be questioned by someone in 6 months Examples: choosing a database, adopting a framework, changing auth strategy, restructuring the API, adding a build tool. ## Template Create files in `docs/decisions/` (or `adr/`) with sequential numbering: ```markdown # ADR-001: Use PostgreSQL for primary database ## Status Accepted | Proposed | Deprecated | Superseded by ADR-XXX ## Date 2026-04-10 ## Context What is the problem or situation that requires a decision? Include constraints, requirements, and forces at play. ## Options Considered ### Option A: PostgreSQL - Pros: ACID compliance, JSON support, mature ecosystem, free - Cons: Requires managing connections, vertical scaling limits ### Option B: MongoDB - Pros: Flexible schema, horizontal scaling - Cons: No transactions across collections, eventual consistency issues ### Option C: PlanetScale (MySQL) - Pros: Serverless, branching, managed - Cons: Vendor lock-in, no foreign keys in their branching model ## Decision We chose **PostgreSQL** because: 1. Our data is relational — users, teams, projects with clear relationships 2. We need ACID transactions for billing operations 3. JSON columns give us schema flexibility where needed 4. Prisma/Drizzle have excellent Postgres support ## Consequences - We need to manage connection pooling (use PgBouncer or Prisma's built-in pool) - Migrations must be backwards-compatible for zero-downtime deploys - We accept vertical scaling limits and will shard later if needed ``` ## Workflow 1. **Identify** the decision being made 2. **Research** the options (at least 2-3 alternatives) 3. **Write** the ADR using the template 4. **Review** with the team (PR or discussion) 5. **Merge** as accepted 6. **Reference** from code comments when relevant: `// See ADR-001` ## File Naming ``` docs/decisions/ ├── 001-use-postgresql.md ├── 002-adopt-trpc-over-rest.md ├── 003-switch-to-pnpm.md └── template.md ``` ## Tips - Keep ADRs short — 1-2 pages max - Write them in the present tense ("We choose X" not "We chose X") - It's OK to write an ADR after the fact if you forgot to write one during the decision - Superseded ADRs should link to their replacement, not be deleted - ADRs are not design docs — they capture the decision, not the full design
Comments (0)
Sign in to join the conversation.
Reviews (0)
No reviews yet.
No comments yet.