skills-web-dev
Skill25 starsChanged 2 months agoMITView on GitHub
Top-level router for the Chwezi Core Systems web-development and software-engineering skills engine. Use for AI systems, SaaS, architecture, APIs, databases, security, frontend engineering, mobile, DevOps, reliability, product engineering, SDLC documentation, catalog maintenance, delivery evidence packs, routing fixtures, and world-class engineering quality gates.
In this repository4 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .github/copilot-instructions.mdCopilot instructions
- 00-meta-initializationSkill
The file
We split this skill into the rules it gives an agent
8 rules. About 0 are common, 2 are distinctive, and 6 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
Always apply these before domain work:
- Very specific: line 25
When a task requires external research, use the Digital Research Engine at C:\wamp64\www\digital-research-engine, beginning with its …
- Very specific: line 52
Use docs/world-class-exemplars/running-example.md: a Chwezi Core Systems multi-tenant field-service SaaS called FieldOps Ledger.
- Very specific: line 56
Read the selected domain SKILL.md.
- Distinctive: line 58
If current facts are needed, check the appropriate source register and invoke Digital Research Skills Engine verification.
- Very specific: line 60
Run python -X utf8 scripts\skill_catalog_guardrails.py --report-only after routing or catalog changes.
- Very specific: line 61
Run python -X utf8 scripts\routing_smoke_test.py --report-only after routing fixtures or frontmatter changes.
- Very specific: line 66
Stop and ask or research further when the request lacks a decision/audience, requires current platform/legal/statutory facts that have not …
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
Skills Web Dev Router
This is the entrypoint for the skills-web-dev engine. It routes engineering work to the smallest accurate active skill, applies quality gates, and records evidence so outputs can be reviewed by a team that did not write them.
Standard implemented: Skills Web Dev July 2026 upgrade baseline. Last verified: 2026-07-08.
Non-negotiable foundation
Always apply these before domain work:
skills/sdlc-meta/anti-ai-slop/SKILL.mdwhile producing any artifact.skills/sdlc-meta/ai-slop-audit/SKILL.mdafter each major iteration and at release.templates/delivery-dod/evidence-pack.mdfor implementation, architecture, security, data, DevOps, AI, SaaS, mobile, or documentation deliverables.docs/source-registers/ai-platforms.mdwhen work depends on current AI, Apple, cloud, security, or framework facts.skills/sdlc-meta/kaizen-improvement-system/SKILL.mdfor every engine audit, product audit, book-driven improvement, and post-iteration learning cycle. Published audits are capped at 65/100; improvement plans target 95/100 and require evidence.
When a task requires external research, use the Digital Research Engine at C:\wamp64\www\digital-research-engine, beginning with its source-evaluation and source-verification skills.
Route by intent
| Intent | Primary skill family |
| AI application, model gateway, RAG, evaluations, agent runtime, HITL, AI cost or security | skills/ai/* |
| API contract, system architecture, distributed systems, microservices, validation contracts | skills/architecture/* |
| Database schema, SQL, MySQL, PostgreSQL, reliability, persistence | skills/backend-databases/* |
| CI/CD, cloud, containers, Kubernetes, release engineering, observability, reliability | skills/devops-cloud/* |
| React, Next.js, Tailwind implementation, frontend performance, image compression, content UX | skills/frontend-ux/* |
| Android, iOS, Kotlin Multiplatform, PWA, mobile operations | skills/android/*, skills/ios/*, skills/mobile-cross/* |
| Desktop applications, Avalonia UI and Python executable-suite packaging | skills/frontend-ux/avalonia-desktop-development, skills/languages/python-modern-standards, skills/languages/csharp-dotnet-development |
| GIS maps, spatial data, PostGIS and mapping APIs | skills/gis/gis-platform-engineering |
| Games, gameplay, interactive narrative, game assets, builds and live operations | skills/game-development/game-development-orchestration then the required specialist |
| TypeScript, JavaScript, Python, PHP, C#/.NET, Java/JVM, Spring, Jakarta EE, Hibernate, Maven, Gradle, WebLogic, Quarkus, Micronaut, Helidon, GraalVM | skills/languages/* |
| SaaS tenancy, pricing, billing, entitlements, SSO/SCIM, portability, admin tooling | skills/saas/* |
| Security audit, code safety, DPIA, Linux hardening, network security | skills/security/* |
| Product discovery, business metrics, documents, spreadsheets, proposal readiness, delivery control | skills/product-business/* |
| SDLC documentation, skill writing, catalog guardrails, world-class engineering gates | skills/sdlc-meta/* and 00-meta-initialization/* |
| Finance/accounting rules, statutory values, IFRS, close, audit, payroll, tax | External chwezi-accounting-doctrine; local finance skills only orchestrate implementation |
| Visual design, typography, UI/UX appearance, documents, slides, spreadsheets, PDF visual polish | External design-system-skills |
| Complete website delivery, content, SEO and launch | External website-skills; retain technical implementation here |
| Formal SRS, lifecycle specifications and governance artefacts | External srs-skills; retain implementation evidence here |
Running example
Use docs/world-class-exemplars/running-example.md: a Chwezi Core Systems multi-tenant field-service SaaS called FieldOps Ledger. The same example threads through architecture, API, database, identity, security, reliability, deployment, and evidence-pack artifacts.
Release workflow
- Read the selected domain
SKILL.md. - Load only the domain references, templates, and examples that the skill points to.
- If current facts are needed, check the appropriate source register and invoke Digital Research Skills Engine verification.
- Produce the artifact and evidence pack together.
- Run
python -X utf8 scripts\skill_catalog_guardrails.py --report-onlyafter routing or catalog changes. - Run
python -X utf8 scripts\routing_smoke_test.py --report-onlyafter routing fixtures or frontmatter changes. - Ship only when the evidence pack, anti-slop gate, and relevant domain validation pass.
Stop conditions
Stop and ask or research further when the request lacks a decision/audience, requires current platform/legal/statutory facts that have not been verified, asks for fabricated evidence, changes active-skill count without a routing reason, or would delete/move skill directories contrary to AGENTS.md.
See also
README.mddocs/skill-routing-index.mddocs/skill-aliases.ymldocs/catalog-cleanup/empty-directory-disposition.mdtemplates/delivery-dod/evidence-pack.mdexamples/full-stack-saas-reference/README.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.