architecture-decision-records
Skill2 starsChanged 9 months agoMITView on GitHub
Use when making architectural decisions during planning or design sessions, when choosing between technologies, patterns, or system designs that affect multiple components or are hard to reverse
66 other projects use a version of this.
Adopted by
Most recently taken up by
- Rojios/everything-claude-codetheir own version, first seen Nov 8, 2026
- kinhluan/skillstheir own version, first seen Nov 8, 2026
- GSA-TTS/devCrew_s1their own version, first seen Nov 8, 2026
- lhbsaa/apex-unifiedtheir own version, first seen Nov 8, 2026
- Lukk17/agent-standardstheir own version, first seen Nov 7, 2026
- pankaj-mahaur/Agent-Daemontheir own version, first seen Nov 6, 2026
Used with
Projects with it also use nuxt4-patterns, pytorch-patterns 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.
loulanyueSkillnuxt4-patterns
Nuxt 4 app patterns for hydration safety, performance, route rules, lazy loading, and SSR-safe data fetching with useFetch and useAsyncData.
Also in 7 projects that use this
Yog-SothoSkillpytorch-patterns
PyTorch deep learning patterns and best practices for building robust, efficient, and reproducible training pipelines, model architectures, and data loading. Always activate when the user is writing PyTorch models or training scripts, reviewing deep learning code, debugging training loops or data pipelines, optimizing GPU memory or training speed, setting up reproducible experiments, implementing transfer learning or fine-tuning, or asks anything about nn.Module, DataLoader, autograd, AMP, or torch.compile.
Also in 10 projects that use this
royaluniondesign-sysSkillcontent-hash-cache-pattern
Cache expensive file processing results using SHA-256 content hashes — path-independent, auto-invalidating, with service layer separation.
Also in 10 projects that use this
junimnjwSkillralphinho-rfc-pipeline
RFC-driven multi-agent DAG execution pattern with quality gates, merge queues, and work unit orchestration.
Also in 7 projects that use this
junimnjwSkillcarrier-relationship-management
Codified expertise for managing carrier portfolios, negotiating freight rates, tracking carrier performance, allocating freight, and maintaining strategic carrier relationships. Informed by transportation managers with 15+ years experience. Includes scorecarding frameworks, RFP processes, market intelligence, and compliance vetting. Use when managing carriers, negotiating rates, evaluating carrier performance, or building freight strategies.
Also in 6 projects that use this
royaluniondesign-sysSkilldeployment-patterns
Deployment workflows, CI/CD pipeline patterns, Docker containerization, health checks, rollback strategies, and production readiness checklists for web applications.
Also in 18 projects that use this
- +2 more from junimnjw
Other versions66 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 66
- loulanyue/awesome-claude-notes★ 272same name
- tranhieutt/software_development_department★ 71same name
- Fmarzochi/EGC★ 61same name
- GSA-TTS/devCrew_s1★ 51same name
- kumaran-is/claude-code-onboarding★ 35same name
- ricoberger/dotfiles★ 19same name
- listener-He/java-harness-agent★ 16same name
- woxiangyangzhimao/skills★ 14same name
- xfstudio/skills★ 9same name
- 4gentic/obelus★ 8same name
56 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 repository11 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- boy-scout-ruleSkill
- domain-driven-designSkill
- functional-core-imperative-shellSkill
- git-commit-messageSkill
- react-best-practicesSkill
- revise-for-claritySkill
- ux-design-principlesSkill
- ux-prototypingSkill
- ux-visual-evaluationSkill
- visual-feedback-loopSkill
The file
We split this skill into the rules it gives an agent
6 rules. 0 are common, 3 are distinctive, and 3 are very specific to this project.
0
Common
Most of the 62 edited versions we read keep these.
3
Distinctive
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
Do NOT interrupt the planning flow to write ADRs.
- Distinctive: line 56
Keep it to these four sections.
- Very specific: line 71
Use the prompt template in proposer-prompt.md.
- Very specific: line 79
Use the prompt template in reviewer-prompt.md.
- Distinctive: line 90
If the reviewer flags issues, the proposer revises and the reviewer re-reviews.
- Very specific: line 94
If a later session revisits a decision, don't edit the original.
“Common” is measured: kept by at least 6 in 10 of the 62 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
Architecture Decision Records
Capture significant architectural decisions as lightweight documents during planning and design. Uses Michael Nygard's ADR template with a two-subagent workflow (propose → review).
When to Use
During planning or design sessions, when architectural decisions are being made. Create ADRs after the planning conversation concludes, before transitioning to implementation.
Do NOT interrupt the planning flow to write ADRs. Collect qualifying decisions as they emerge, then batch-create ADRs at the end.
Threshold Gate
Apply this checklist to each decision. If 2 or more are yes → write an ADR.
- Does it affect multiple components or services?
- Does it constrain future technical choices?
- Does it involve choosing between meaningfully different alternatives?
- Would it be difficult or costly to reverse?
NOT worth an ADR: utility library choices, naming conventions, code style, implementation details within a single module, anything already decided by an existing ADR.
The line: ADRs capture decisions that shape the system's structure, not decisions that shape a single file.
Template (Michael Nygard)
Store in docs/decisions/NNN-short-title.md. Number sequentially (001, 002, ...). Title uses present-tense imperative verb phrase (e.g., use-postgres-for-persistence).
# NNN. Short Title as Present-Tense Imperative Verb Phrase
## Status
Proposed | Accepted | Deprecated | Superseded by [NNN]
## Context
What is the issue? What forces are at play? State facts and constraints,
not opinions. Include technical constraints, project requirements, and
team factors.
## Decision
We will [chosen approach].
## Consequences
What becomes easier or harder? Be honest — include both positive and
negative consequences.Keep it to these four sections. Do not add extra sections like "Alternatives Considered" or "Deciders" — alternatives belong in Context, and the git log records who decided.
Status Rules
- Human-directed decisions (the human explicitly chose during the session): Status =
Accepted. The human already decided; the ADR documents it. - Agent-initiated decisions (the agent made an architectural choice not explicitly discussed with the human): Status =
Proposed. Surface to human for approval before implementation proceeds.
Workflow
1. Collect decisions during planning
As architectural choices emerge during planning or design, note which ones cross the threshold. Track whether each was human-directed or agent-initiated.
2. Dispatch proposer subagent (one per decision)
Use the prompt template in proposer-prompt.md. The proposer:
- Checks existing ADRs in
docs/decisions/to determine the next number and avoid contradictions - Writes the ADR using the Nygard template
- Sets status per the rules above
- Saves the file and commits
3. Dispatch reviewer subagent
Use the prompt template in reviewer-prompt.md. The reviewer checks:
- Is the Context factual and complete?
- Is the Decision stated clearly?
- Are Consequences honest (both positive AND negative)?
- Does this actually warrant an ADR (threshold re-check)?
- Does it contradict existing ADRs?
Reports: Approved or Issues with specific feedback.
4. Loop if needed
If the reviewer flags issues, the proposer revises and the reviewer re-reviews.
5. Updating existing ADRs
If a later session revisits a decision, don't edit the original. Create a new ADR and set the old one's status to Superseded by [NNN].
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.

