459 tagged software-engineering, measured the same way as everything else here.
Browse within: software-development 105agentic-coding 83coding-agents 70metaprogramming 60analyzing 48artificial-intelligence 48codex-plugin 47software-architecture 45skill-pack 38subagents 38evidence-based 35agentic-workflow 29Developer Workflow 20context-engineering 19
tsai09495/matt-pocock-engineering
Skill Claude CodeCodex
Review working-tree, branch, commit, or PR changes on separate Standards and Spec axes. Use for code review, uncommitted-change inspection, or checking a diff against source requirements.
tsai09495/matt-pocock-engineering
Skill Claude CodeCodex
Challenge and sharpen a plan against repository evidence, domain language, and ADRs. Use for a dependency-safe decision interview or when assumptions must be resolved before a spec or implementation.
tsai09495/matt-pocock-engineering
Skill Claude CodeCodex
Map a huge, foggy engineering effort into durable decision tickets. Use only when the user explicitly selects Wayfinder for a multi-session destination with dependent unknowns.
Skill Claude CodeCodex
Fix all guardrail findings (make lint, make test, make sast) across repositories. Use when the user asks to fix linting errors, test failures, SAST findings, or run the full guardrail remediation workflow.
Skill Claude CodeCodex
Scaffold a new frontend project with 5-layer Clean Architecture (Domain, Service, Infrastructure, Presentation, Main). Use when the user asks to create, scaffold, or bootstrap a new frontend or React project.
Skill Claude CodeCodex
Scaffold a new Go backend project with Clean Architecture, Dig DI, and testify. Use when the user asks to create, scaffold, or bootstrap a new Go project.
sarveshtalele/mcp-skills-registry
Skill Claude CodeCodex
Perform a deterministic, graph-driven Change Impact Analysis for a repository. Given a set of changed files, builds a dependency graph, finds directly and transitively affected modules, validates API contracts for breaking changes, parses CODEOWNERS, and computes a 0-100 deployment risk score. Trigger on: change…
sarveshtalele/mcp-skills-registry
Skill Claude CodeCodex
Reverse-engineer / discover a codebase. Works in two modes — provide EXACTLY one input: repourl (a github.com URL) clones the repo server-side and runs full static analysis, producing a System Design Document, architecture + dependency overview, and a 100-point quality score; repopath (a local directory the server can…
joelbrilliant/agentic-delivery-skills
Skill Claude CodeCodex
Use when reviewing code adversarially or building production-bound features where rigour matters more than speed. Required for code reviews ending in a verdict (PASS/HOLD), when verifying that "tests pass" actually proves a feature works end-to-end, before closing findings as non-issues, before signing off your own…
joelbrilliant/agentic-delivery-skills
Skill Claude CodeCodex
Preserves useful state across long agent sessions, context compaction, handoffs, and restarts. Use when compressing context, handing work between agents, resuming after interruption, writing session summaries, or preventing repeated work.
joelbrilliant/agentic-delivery-skills
Skill Claude CodeCodex
Enforces a design/spec approval gate before production implementation. Use when UI, UX, product flows, architecture, API shape, prototypes, visual direction, or user-facing behaviour need signoff before builders write production code.
william-london/ownframework-loop
Skill Claude CodeCodex
Part of of-loop
OwnFramework Loop - one bounded build or repair pass. Claims atomically, consumes deterministic preparation, materializes a pass-scoped semantic skeleton, invokes a fresh of-builder, and deterministically finalizes.
william-london/ownframework-loop
Skill Claude CodeCodex
Part of of-loop
OwnFramework Loop - one bounded review pass. Waits until a build receipt exists, then performs exact-SHA review and transitions state via the deterministic review finalizer.
william-london/ownframework-loop
Skill Claude CodeCodex
Part of of-loop
OwnFramework Loop - create, inspect, amend, or stop a run. Returns the supervisor-first unattended handoff plus optional foreground/debug commands.
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: