adr
Skill1 starsChanged 2 months agoMITView on GitHub
Creates and maintains Architecture Decision Records (ADRs) in docs/decisions/ using the MADR 4.0 format. Supports creating, listing, updating, superseding, and searching ADRs.
90 other projects use a version of this.
Adopted by
Most recently taken up by
- GustavoAlecio/claude-setuptheir own version, first seen Nov 8, 2026
- computerlovetech/skillstheir own version, first seen Nov 8, 2026
- rvdbreemen/OTGW-firmwaretheir own version, first seen Nov 8, 2026
- axodouble/skillstheir own version, first seen Nov 8, 2026
- RodrigoAnciaes/golem-sddtheir own version, first seen Nov 8, 2026
- n1kben/agentstheir own version, first seen Nov 8, 2026
Used with
Projects with it also use fdr, runbook and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
thiagofgfSkillfdr
Use when creating, changing, reviewing, retiring, auditing, or backfilling implemented user-visible feature behavior, permissions, feature-specific rationale, tradeoffs, or relationships to architectural decisions.
Also in 3 projects that use this
AnthropicSkillrunbook
Create or update an operational runbook for a recurring task or procedure. Use when documenting a task that on-call or ops needs to run repeatably, turning tribal knowledge into exact step-by-step commands, adding troubleshooting and rollback steps to an existing procedure, or writing escalation paths for when things go wrong.
Also in 5 projects that use this
drewdrewthisSkillsweep
Search for similar patterns, anti-patterns, or issues elsewhere in the codebase after a fix or feature. Use after implementing a fix to catch the same bug in other places, or proactively to audit for a known pattern.
Also in 4 projects that use this
vndeeSkillproduct-spec
Use when starting a new feature, product, or project — before any architecture or code, to define what to build and why
Also in 3 projects that use this
OpenAISkillthreat-model
Use when Codex is already in the threat-modeling phase of a security scan, the user explicitly invokes $threat-model, or the user explicitly asks to create, update, or persist a repository threat model. Do not use as the primary trigger for full PR, commit, branch, patch, or repository scans.
Also in 6 projects that use this
montymiSkilltech-spec
Parse and review Technical Specification documents for completeness, consistency, and implementation readiness
Also in 3 projects that use this
Other versions90 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 90
- CrowdDotDev/crowd.dev★ 3.4ksame name
- chattocorp/chatto★ 2.9ksame name
- Cosmian/kms★ 372same name
- linuxfoundation/insights★ 288same name
- rvdbreemen/OTGW-firmware★ 213same name
- sd0xdev/sd0x-dev-flow★ 192same name
- sd0xdev/sd0x-harness★ 192same name
- irahardianto/awesome-agv★ 157same name
- jellydn/my-ai-tools★ 127same name
- EmeaAppGbb/spec2cloud★ 100same name
80 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 repository22 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- accessibilitySkill
- api-designSkill
- architecture-designSkill
- architecture-reviewSkill
- best-practicesSkill
- changelog-generatorSkill
- code-reviewSkill
- database-designSkill
- dependency-auditSkill
- doc-refreshSkill
- example-skillSkill
- full-security-reviewSkill
- git-workflowSkill
- golden-rulesSkill
- incident-reportSkill
- log-correlationSkill
- mvp-readinessSkill
- onboardingSkill
- pre-commitSkill
- requirements-generatorSkill
- test-writerSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
13 rules. 0 are common, 8 are distinctive, and 5 are very specific to this project.
0
Common
Most of the 75 edited versions we read keep these.
8
Distinctive
This author's own choices. Worth a look.
5
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
Create, list, update, and supersede Architecture Decision Records (ADRs) following the Markdown Architectural Decision Records (MADR) 4.0 …
- Distinctive: line 106
If no arguments were given, ask the user what they want to do:
- Distinctive: line 117
Use Glob to find all existing ADRs in the detected ADR directory.
- Distinctive: line 118
Use Read on any related existing ADRs mentioned.
- Distinctive: line 119
Ask the user (skip the title question if it was passed as an argument):
- Very specific: line 127
Use Write to save to <adr-dir>/ADR-<NNN>-<kebab-case-title>.md.
- Distinctive: line 128
Update the ADR index (see Step 3).
- Very specific: line 134
Use Glob to find all ADRs in the detected ADR directory.
- Very specific: line 150
Use Read to load the ADR. Use Edit to update the status front matter key (or the legacy Status: line in older ADRs) and set date to today.
- Distinctive: line 161
Create a new ADR (Step 2a) that documents the new decision.
- Very specific: line 163
Use Edit to update the old ADR's status to superseded by ADR-NNN.
- Distinctive: line 169
Use Grep across the detected ADR directory for the keyword or tag.
- Very specific: line 175
After any create/update operation, regenerate <adr-dir>/README.md — an index of all ADRs sorted by number, showing status as a text badge: …
“Common” is measured: kept by at least 6 in 10 of the 75 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
adr
Purpose
Create, list, update, and supersede Architecture Decision Records (ADRs) following the Markdown Architectural Decision Records (MADR) 4.0 format. Maintains an ADR directory (see Directory Convention below). Works standalone or as a companion to /architecture-design.
Directory Convention
Before creating anything, detect where this project keeps ADRs. Use Glob to check, in order: docs/decisions/ADR-*.md, docs/decisions/[0-9][0-9][0-9][0-9]-*.md, docs/adr/*.md, adr/*.md, docs/architecture/decisions/*.md. Use the first directory that already contains ADRs. Only if none exists, default to docs/decisions/. All paths below refer to this detected directory. If existing ADRs use a different file-name pattern (e.g. MADR's NNNN-title.md), follow that pattern instead of ADR-NNN-title.md.
ADR Format (MADR 4.0)
Every ADR follows the MADR 4.0.0 template (https://adr.github.io/madr/). Metadata lives in YAML front matter. tags is not part of MADR; this skill adds it as an optional key so list and search can filter by area. Sections marked <!-- optional --> may be dropped when they add nothing; remove the marker comments from the saved file.
---
status: proposed | accepted | rejected | deprecated | superseded by ADR-NNN
date: YYYY-MM-DD
decision-makers: <names or roles>
consulted: <people whose opinions were sought; two-way communication>
informed: <people kept up to date; one-way communication>
tags: <architecture | security | database | api | infrastructure | ...>
---
# ADR-NNN: <Title>
## Context and Problem Statement
[2-3 sentences: what situation led to this decision? What problem is being solved? Phrase it as a question if that helps.]
<!-- optional -->
## Decision Drivers
* [driver 1: most important constraint or goal]
* [driver 2]
## Considered Options
* [Option A]
* [Option B]
* [Option C]
## Decision Outcome
Chosen option: "[Option A]", because [justification tied to the decision drivers].
<!-- optional -->
### Consequences
* Good, because [what becomes easier, better, or possible]
* Bad, because [what becomes harder, more expensive, or is given up]
* Neutral, because [effect that is neither a gain nor a loss]
<!-- optional, but include it whenever possible -->
### Confirmation
[How compliance with this decision will be checked: design or code review, an architecture test (e.g. ArchUnit), a lint rule, a CI check.]
<!-- optional -->
## Pros and Cons of the Options
### [Option A]
* Good, because [argument]
* Neutral, because [argument]
* Bad, because [argument]
### [Option B]
* Good, because [argument]
* Bad, because [argument]
<!-- optional -->
## More Information
[Supporting evidence, team agreement, when to revisit the decision, links to related ADRs, design docs, RFCs, or tickets.]Instructions
Step 1 — Determine the action
Parse $ARGUMENTS first and dispatch directly — do not ask if the action is given:
| Invocation | Action |
/adr create <title> | Step 2a, using <title> as the decision title |
/adr list | Step 2b |
/adr update <n> <status> | Step 2c on ADR-<n>, setting <status> |
/adr supersede <n> | Step 2d replacing ADR-<n> |
/adr search <keyword> | Step 2e with <keyword> |
If no arguments were given, ask the user what they want to do:
- (a) Create a new ADR — document a new architectural decision
- (b) List existing ADRs — show all ADRs with status and one-line summary
- (c) Update an ADR's status — mark as Accepted, Rejected, Deprecated, or Superseded
- (d) Supersede an ADR — create a new ADR that replaces an existing one
- (e) Search ADRs — find ADRs by tag, status, or keyword
Step 2a — Create a new ADR
- Use Glob to find all existing ADRs in the detected ADR directory. Determine the next number.
- Use Read on any related existing ADRs mentioned.
- Ask the user (skip the title question if it was passed as an argument):
- What is the decision being made? (one sentence title)
- What is the context / problem? (what forced this decision)
- Who made the decision? Who was consulted, and who needs to be informed?
- What options were considered?
- What was chosen and why?
- How will compliance be confirmed (review, test, CI check)?
- Generate the full ADR using the MADR format above.
- Use Write to save to
<adr-dir>/ADR-<NNN>-<kebab-case-title>.md. Create the directory if it doesn't exist. - Update the ADR index (see Step 3).
Step 2b — List existing ADRs
Use Glob to find all ADRs in the detected ADR directory. Use Read on each to extract: number, title, status, date, tags. Read these from the YAML front matter; for older ADRs that use bold **Status:**-style fields instead, read those. Output a sorted table:
## Architecture Decision Records
| # | Title | Status | Date | Tags |
|---|-------|--------|------|------|
| ADR-001 | Use PostgreSQL as primary database | Accepted | 2026-01-15 | database |
| ADR-002 | Adopt hexagonal architecture | Accepted | 2026-01-20 | architecture |
| ADR-003 | Use Redis for session storage | Proposed | 2026-02-01 | infrastructure |Step 2c — Update status
Use Read to load the ADR. Use Edit to update the status front matter key (or the legacy **Status:** line in older ADRs) and set date to today. Valid transitions:
- Proposed → Accepted (decision confirmed)
- Proposed → Rejected (decision considered and turned down)
- Proposed → Deprecated (no longer relevant before being accepted)
- Accepted → Deprecated (no longer the right approach)
- Accepted → Superseded by ADR-NNN (replaced by a newer decision)
Step 2d — Supersede
- Create a new ADR (Step 2a) that documents the new decision.
- In the new ADR's More Information section, add a link: "Supersedes ADR-NNN: [title]"
- Use Edit to update the old ADR's
statustosuperseded by ADR-NNN.
Step 2e — Search
Use Grep across the detected ADR directory for the keyword or tag. Return matching ADRs with context.
Step 3 — Maintain ADR Index
After any create/update operation, regenerate <adr-dir>/README.md — an index of all ADRs sorted by number, showing status as a text badge: [ACCEPTED], [PROPOSED], [REJECTED], [DEPRECATED], [SUPERSEDED].
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.

