Memory Decide

VerifiedSafe

Creates Architecture Decision Records (ADRs) to document significant technical choices. Use it when you need to record the rationale behind a design or architecture decision, including context, options considered, and the chosen approach. It stores records in a structured markdown format in the project's memory system for future reference.

Sby Skills Guide Bot
DevelopmentIntermediate
6/2/2026
Claude Code
#architecture-decision-records#documentation#decision-making#adr#technical-decisions

Recommended for

Automated summary

Creates Architecture Decision Records (ADRs) to document important architectural and design decisions in a structured format.

Strengths

  • Clear, repeatable structure for documenting decisions
  • Numbering and slug system for easy retrieval
  • Supports quick inline input mode
  • Encourages consideration of alternatives and consequences

Limitations

  • Requires discipline to keep records up to date
  • Does not automatically handle retroactive changes or superseded decisions
  • Depends on user providing complete information
When to use it

Use this skill when making a significant technical decision that needs to be tracked and referenced later.

When not to use it

Avoid using it for trivial choices or quick notes that do not require formal documentation.

Security analysis

Safe
Quality score92/100

The skill only reads and writes files within a designated documentation directory (.claude/memory/decisions/). It does not execute any external commands, network calls, or destructive actions. File path sanitization (lowercase, underscore replacement, removal of special characters) reduces risk of path traversal. No sensitive data is exfiltrated. The operations are purely for creating decision records based on user input.

No concerns found

How this analysis is made

Examples

Create a new ADR for database choice
I need to document our decision to use PostgreSQL as the primary database. We considered MySQL and MongoDB but chose PostgreSQL for its advanced indexing and ACID compliance.
Quick mode inline ADR
/memory-decide Use JWT for API auth because it's stateless and works well with microservices. Considered session cookies but they don't scale across services.

description: Create decision records (ADRs) to document architectural and design decisions. Use when making significant technical choices.

Memory Decide

Create a new decision record (ADR) in the memory system.

Instructions

When this skill is triggered, perform the following steps:

Step 1: Find Next Decision Number

Scan the .claude/memory/decisions/ directory for existing decision files.

Decision files follow the format: NNNN_decision_slug.md Examples:

  • 0001_database_choice.md
  • 0002_authentication_strategy.md

Find the highest existing number and increment by 1. If no files exist, start at 0001.

Step 2: Prompt for Decision Details

Ask the user for the following information:

Required:

  1. Decision Title: Brief title describing the decision
  2. Context: What problem or situation led to this decision?
  3. Options Considered: What alternatives were evaluated?
  4. Chosen Option: Which option was selected and why?

Optional: 5. Consequences: What changes as a result of this decision? 6. Status: Accepted (default), Superseded, or Deprecated

Example prompt:

To create a decision record, please provide:

1. **Title** (required): What is this decision about?
   Example: "Use PostgreSQL for primary database"

2. **Context** (required): What problem are we solving?

3. **Options** (required): What alternatives did you consider?
   Format each option with pros/cons if possible

4. **Decision** (required): What did you choose and why?

5. **Consequences** (optional): What changes as a result?

Step 3: Generate Slug

Create a URL-friendly slug from the title:

  • Convert to lowercase
  • Replace spaces with underscores
  • Remove special characters
  • Limit to 50 characters

Step 4: Create Decision File

Create the file at .claude/memory/decisions/NNNN_slug.md using this template:

# Decision NNNN: [Title]

**Date**: YYYY-MM-DD
**Status**: Accepted

## Context
[What issue are we facing? What motivated this decision?]

## Options Considered

### Option 1: [Name]
[Description]
- **Pros**: [advantages]
- **Cons**: [disadvantages]

### Option 2: [Name]
[Description]
- **Pros**: [advantages]
- **Cons**: [disadvantages]

## Decision
[What we chose and the rationale behind the choice]

## Consequences
[What changes as a result of this decision]
- [Positive consequence]
- [Negative consequence or trade-off]
- [Required follow-up actions]

Step 5: Confirm Creation

Display confirmation to user:

## Decision Record Created

**File**: `.claude/memory/decisions/NNNN_slug.md`
**Title**: Decision NNNN: [Title]
**Date**: YYYY-MM-DD
**Status**: Accepted

### Summary
[Brief summary of the decision]

### Options Evaluated
1. [Option 1 name]
2. [Option 2 name]

### Chosen: [Selected option]

---
Decision recorded. Reference this as "Decision NNNN" in future discussions.

File Paths

  • Decisions directory: .claude/memory/decisions/
  • File format: NNNN_slug.md (zero-padded 4-digit number)

Error Handling

  • If .claude/memory/decisions/ doesn't exist, create it
  • If user provides incomplete info, ask follow-up questions
  • If slug collision occurs, append number: 0005_api_design_2.md

Quick Decision Mode

If user provides all info inline, skip the prompts and create directly:

User: /memory-decide Use JWT for API auth because it's stateless and works well with microservices. Considered session cookies but they don't scale across services.

Claude parses the input and creates the record immediately.

From the file to a first result

The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.

View the source and supporting files

This download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.

1. Download the file

Download SKILL.md

From the root of a test project, save the file as .skill-downloads/memory-decide-mm78nnyy/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.

Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.

mkdir -p '.skill-downloads/memory-decide-mm78nnyy' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/memory-decide-mm78nnyy/download' --output '.skill-downloads/memory-decide-mm78nnyy/SKILL.md.part' && mv '.skill-downloads/memory-decide-mm78nnyy/SKILL.md.part' '.skill-downloads/memory-decide-mm78nnyy/SKILL.md'

2. Check prerequisites

Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.

No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.

The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.

3. Load the skill

This file does not provide a usable name/description header for this native procedure. It remains available to download and inspect; use the manual method below.

Manual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.

4. Try a first task

Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:

Read .skill-downloads/memory-decide-mm78nnyy/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.

Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.

Related skills