Evaluate a developer's engineering problem and recommend the best approach with a score. Use this skill whenever a developer describes an issue they need to solve and optionally proposes an initial idea — the typical case is something like "we have a problem with X, should we do Y?" or "I need to solve X, my idea is…
Decompose a feature specification (.md file) into a structured roadmap of goals and tasks, written to /tmp/FEATURE-Manager-{projectname}-{timestamp}.md. Use this skill whenever the user provides a markdown spec describing a feature, product, or system and asks to "break it down", "decompose it", "create goals and…
Apply the blkst8 Go codebase style guide when generating, scaffolding, or reviewing Go code. ALWAYS use this skill when the user asks to create a new Go project or service, scaffold any Go code, add a handler / repository / service / model / worker / migration, set up HTTP routing with Echo, configure logging (zap) or…
Apply the blkst8 Go codebase style guide when generating, scaffolding, or reviewing Go code. ALWAYS use this skill when the user asks to create a new Go project or service, scaffold any Go code, add a handler / repository / service / usecase / model / worker / migration, set up HTTP routing with Echo, configure…
Self-improving agent skill — ALWAYS ACTIVE on every task execution. After every task the agent observes, reflects, extracts patterns, scores them, generates skills, and registers them autonomously. Explicitly trigger when the user says "self-heal", "auto-skill", "learn from this", "create a skill for this", "add this…
Review and validate agent specification (.md) files. Use this skill whenever the user provides a .md file containing an agent workflow, prompt template, or behavioral specification and asks you to review it, check it, validate it, improve it, or make sure it's correct. Also use when the user says things like "go over…
Always-on CAPS emphasis skill for reasoning and planning. Use this skill whenever the user asks to highlight CRITICAL constraints, goals, deadlines, budgets, limits, or requirements in ALL CAPS, or asks for a PRIORITY REGISTER / ACTIVE PRIORITIES list across steps. Also trigger when prompts include words like must…
Scaffold, restructure, audit, or work within a frontend + backend monorepo using the apps/ + packages/ layout. Default stack is pnpm workspaces + Turborepo for JS/TS repos, with adaptations for polyglot repos where the backend is Go, Python, or another non-JS language. Use this skill whenever the user mentions a…
Build, test, and improve programs that use typed decision models such as Laya, Jev, or any compatible choice/score/noul backend. Use this skill whenever the user asks to classify, score, gate, route, extract, triage, moderate, or make other bounded judgments from text or structured state, especially when they mention…
Build lean, opinionated products using the 37signals philosophy from "Getting Real", "Rework", and "Shape Up". Use when the user mentions "Getting Real", "Rework", "Shape Up", "37signals", "Basecamp method", "six-week cycles", "fixed time variable scope", "appetite vs estimates", "betting table", "breadboarding", "fat…
Agent skill for code-analyzer - invoke with $agent-code-analyzer.
★not rated 6 14d agoA19 tokens
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: