adr-writer
Skill8 starsChanged 3 months agoMITView on GitHub
Records Architecture Decision Records (ADRs) in the Michael Nygard format. Use when a relevant architectural decision is made (technology choice, structural pattern, significant trade-off) and needs to be persisted with its context and consequences.
12 other projects use a version of this.
Adopted by
Most recently taken up by
- Mathews-Tom/armorytheir own version, first seen Nov 1, 2026
- pathiec92/claude-skillstheir own version, first seen Oct 25, 2026
- llopresto87/Cypresstheir own version, first seen Oct 14, 2026
- mikulgohil/claude-skillstheir own version, first seen Oct 13, 2026
- GabrielHenriqueCA/cortex-aitheir own version, first seen Oct 11, 2026
- govctl-org/govctltheir own version, first seen Oct 9, 2026
Other versions12 projects use a version of this.
Word for word · 1
- jcarlos78/GAIDE★ 8100% same
Edited · 0
None found.
Wrote their own version · 11
- Mathews-Tom/armory★ 340same name
- govctl-org/govctl★ 276same name
- llopresto87/Cypress★ 119same name
- applypack/applypack★ 30same name
- 45ck/software-architecture-skills★ 16same name
- brpaz/agent-skills★ 1same name
- GabrielHenriqueCA/cortex-ai★ 1same name
- kuju63/skills★ 0same name
- mikulgohil/claude-skills★ 0same name
- pathiec92/claude-skills★ 0same name
1 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository7 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .claude/CLAUDE.mdCLAUDE.md
- code-reviewerSkill
- security-reviewerSkill
- spec-writerSkill
- test-generatorSkill
- verifierSkill
The file
Files
<!-- GENERATED from agents/skills/adr-writer.md by scripts/sync-adapters.sh — edit the source and re-run; do not edit this file. -->Skill: adr-writer
When to activate
- Technology choice (database, framework, significant library)
- Structural pattern adopted (authentication, observability, error handling)
- Significant trade-off (chose X over Y because Z)
- A reverted or superseded decision (create a new ADR marking the previous one as
superseded)
Before starting
- List existing ADRs in
docs/adr/. - Determine the next sequential number (4 digits, leading zero).
- Confirm with the user: "I'll record this decision as ADR XXXX. OK?"
Template (adapted Michael Nygard format)
Create file docs/adr/NNNN-title-in-kebab-case.md:
# ADR NNNN — <Short descriptive title>
- **Status:** proposed | accepted | rejected | superseded by ADR XXXX
- **Date:** YYYY-MM-DD
- **Decision-makers:** [names]
## Context
The problem, constraints, requirements. What made this decision necessary. Prior state.
## Decision
What was decided. Present tense: "We will use X."
## Alternatives considered
- **Option A:** pros, cons, reason for rejection.
- **Option B:** ...
## Consequences
### Positive
- ...
### Negative (accepted costs)
- ...
### Neutral
- ...
## Constitution adherence
Which constitution principles this decision honors or tensions.
## Future review
When should this decision be revisited? Under what signals?Principles
- ADR is immutable. Decisions change via a new ADR that marks the previous one
superseded. Do not edit accepted ADRs. - Short is good. An overly long ADR signals an undercooked decision.
- Honest about costs. List the negative consequences candidly; an ADR with no costs is suspicious.
- Present context. Whoever reads this in 2 years needs to understand why this was chosen.
Expected output
One versioned file docs/adr/NNNN-title.md.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

