125 tagged architecture, measured the same way as everything else here.
Browse within: code-quality 55claude-code-plugin 41contracts 31AI Safety 30adr 24ai-development 24browser-automation 24gcp 24developer-tool 19scaffolding 19sdl 19copilot 18ai-generated-code 17coding 16
Agent
Adversarial reviewer for high-impact decisions. Challenges the chosen approach from a clean, unbiased context. Use when /lead or /autonomous-lead triggers a contrarian gate — high impact, hard-to-reverse, or high-uncertainty decisions.
Agent
Systematic code reviewer for .NET/C#. Use proactively after implementation is complete, when delegated from /lead at Phase 9, or when asked to review files, a PR, or a branch. Runs mandatory searches before manual analysis and returns a structured report with an APPROVE / REQUEST CHANGES / NEEDS DISCUSSION verdict.
Agent
End-to-end test expert creating production-quality tests exercising real code with no mocks.
Agent
Creates and manages reusable Agent Skills by analyzing workflows, designing progressive disclosure structure, and generating SKILL.md files with validation. Use when you need to formalize recurring workflows, package team capabilities, create validation skills, or manage existing skills. Triggers on "create skill"…
Agent
Comprehensive framework for creating specialized agents using Claude Code Agent SDK. PROACTIVELY designs, configures, and manages SDK-based agents with precision and consistency.
Agent
Analyze a user task and return a clear, ordered implementation plan before coding.
Agent
Guarantee every screen is operable by keyboard, legible to assistive technology, and free of axe violations — in both LTR and RTL. Accessibility is a release gate here (npm run test:a11y MUST be green), not a nice-to-have, so this agent reviews for the gate, not for vibes.
Agent
Own the custom frontend-architecture ESLint plugin as living law: review violations of the 14 rules, keep the two config-driven maps (layer policy table, package ownership map) in sync with reality, and audit every eslint-disable against its documented exception. Lint runs with --max-warnings=0; a warning is a failure.
Agent
Run the full validation pipeline before anything ships, block on any red gate with zero negotiation, and produce the release notes. The gatekeeper does not review code style or architecture — other agents did that. It verifies that every automated gate is green on the exact commit being released and that the paper…
Agent
Architecture reviewer evaluating system design, SOLID principles, scalability, API design, and technical debt.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: