production-web-architecture
Skill2 starsChanged 4 months agoMITView on GitHub
Design, review, or harden a startup web platform for scalability, reliability, security, observability, deployability, data integrity, background work, caching, AI cost control, and staged growth without premature microservices.
In this repository10 other files it gives its agents.
- AGENTS.mdAGENTS.md
- llms.txtllms.txt
- agentic-site-buildingSkill
- conversion-accessibility-polishSkill
- landing-page-craftSkill
- product-motion-designSkill
- responsive-ui-qaSkill
- technical-seo-discoverySkill
- web-design-systemSkill
- web-visual-directionSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 16
Keep a modular monolith until independent scaling, ownership, deployment, or fault isolation creates a measurable extraction benefit.
- Distinctive: line 18
Add caches only with ownership, key design, TTL, invalidation, stampede protection, and stale behavior defined.
- Distinctive: line 44
Read references/architecture-checklist.md for a production review or extraction decision.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Production Web Architecture
Turn product flows into an operable system. Prefer the smallest architecture that removes the largest demonstrated risk.
Workflow
- Map users, critical journeys, trust boundaries, data ownership, external services, files, asynchronous work, and failure consequences.
- State workload assumptions and service objectives: traffic shape, latency, availability, durability, freshness, privacy, and cost.
- Trace each critical path from browser through API, storage, queue, worker, and response. Mark transaction and idempotency boundaries.
- Review scalability, reliability, security, data integrity, performance, observability, deployment, and cost.
- Keep a modular monolith until independent scaling, ownership, deployment, or fault isolation creates a measurable extraction benefit.
- Move slow or fragile work behind durable queues. Make handlers idempotent, retryable, observable, and visible to users.
- Add caches only with ownership, key design, TTL, invalidation, stampede protection, and stale behavior defined.
- Define migrations, rollout, rollback, health checks, alerts, runbooks, and cost guardrails before calling the design production-ready.
- Validate the highest-risk flow with a failure test, load test, or end-to-end smoke test proportional to the risk.
Startup Defaults
- Managed relational database, object storage, queue, cache, and observability before custom distributed systems.
- Server-side authorization at every ownership boundary.
- Timeouts, pagination, rate limits, and idempotency keys on relevant APIs.
- Structured logs, trace or request IDs, service metrics, and actionable alerts.
- AI calls use structured output, provider routing, token ceilings, prompt versions, and bounded retries.
- Public reads degrade gracefully; writes fail explicitly rather than silently losing data.
Delivery Contract
Provide:
- System map - components, data, trust boundaries, and critical paths.
- Assumptions and SLOs - explicit numbers or identified unknowns.
- Findings - severity, evidence, failure mode, and affected journey.
- Target architecture - smallest justified change and deferred complexity.
- Execution plan - schema, API, queue, migration, rollout, and ownership tasks.
- Verification and operations - tests, dashboards, alerts, rollback, and cost controls.
Reference
Read references/architecture-checklist.md for a production review or extraction decision.
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.

