MarkBovee
UserWorkflow skills and routing for coding agents
Categories
Indexed Skills (32)
agent-workflows
Use when coordinating multi-agent work, parallel execution, task handoff, shared context, or clean session shutdown across multiple agents or terminals. Especially useful when the host supports subagents, hooks, or shared context. Common triggers: multi-agent, parallel work, agent coordination, task handoff, subagent delegation, version bump, bump version, release notes, changelog, tag release, release prep.
code-review
Use when code changed and a meaningful diff is ready; fresh eyes should catch requirement gaps, regressions, or risky design mistakes before handoff or success claims. Common triggers: review, nakijken, pull request, code review, fresh eyes, start reviewing, review deze wijziging, check de wijziging, review changes, second look, bekijk de diff, controleer de code, code check, diff review, PR review.
develop
Default baseline skill for normal software work: small safe iterations, built-in validation, no unnecessary pauses between clear next steps. Includes mode selection (direct, batch, delegate) and cheap-first escalation for bounded mechanical chores. Common triggers: develop, kaizen, autopilot, keep going, continue without waiting, do not stop, don't stop, ga door, werk door, volgende logische stap, volgende stap, start working, start coding, get started, zonder te wachten, niet stoppen, gewoon doorgaan, ga verder, implement this, fix this, add this, pas dit aan, maak dit af, implement, implementeer dit, keep coding, continue implementation, work through steps, code change, start implementing, start implementation, batch edits, delegate work, subagent, maak dit werkend, rewrite, herbouw, coordinator, coördinator, omwerken, entity rewrite, refactor, code aanpassen, werk dit bij, build this, development.
improve
Use when the codebase needs a structured audit, audit-driven plans, execution of those plans, or a focused refactoring/simplification pass. Covers correctness, security, performance, tech debt, migrations, DX, direction, and code cleanup in one skill. Common triggers: improve, audit, tech debt, tech debt audit, audit codebase, improve codebase, direction, audit and plan, refactor this, refactoren, code cleanup, opschonen, simplify this code, vereenvoudigen, remove over-engineering, deduplicate logic, restructure this code, reduce complexity, untangle this, clean architecture mess.
debugging
Use when a bug, failing test, or broken build is not already explained by a clear local mistake, or when a first fix did not work. Common triggers: bug, failing test, broken build, debug, debuggen, error, crash, stack trace, not working, does not work, broke, start debugging, start investigating, fout opsporen, slow startup, timeout, hangt, hanging, crash loop, None, target_temp, niet werkend, doet het niet, malfunction, storing.
intake
Use when the goal, constraints, or success criteria are not yet crisp — from fuzzy design ideas through ambiguous scope to multi-file planning. Covers brainstorming, scoping, and execution planning in one skill. Common triggers: brainstorm, brainstormen, fuzzy idea, design tradeoff, unsure what to build, product direction, idee uitwerken, ambiguous request, unclear scope, behavior-changing work, fuzzy requirements, what should we build, wat moeten we bouwen, wat moeten we maken, best approach, how should we approach this, not sure where to start, start by clarifying, start with questions, ik weet niet waar te beginnen, hoe pakken we dit aan, plan, plannen, multi-file work, multi-phase work, migration, sequencing risk, staged refactor, stages, service by service, per service, dependency chain, sequential steps, per laag, stap voor stap, start planning, start with a plan, werk voorplannen, we moeten dit aanpakken, laten we dit doen, we moeten, laten we, pair programming, samenwerken,...
session-review
Use after completing work to reflect on skill usage, identify gaps, and file improvement issues in the agent-skills-kit repo. Also handles general GitHub issue creation from bug reports, review findings, and follow-ups. Common triggers: retrospective, retro, reflect on session, how did i use skills, file an issue, create issue, github issue, file issue, gh issue create.
verification
Use when about to claim something works, is fixed, or ready to hand off — and when a task changed one or more repos and needs intentional cleanup before stopping. Common triggers: verify, verifiëren, prove, controleren of het werkt, bewijzen dat het werkt, claim success, done, finished, ready, handoff, klaar, gereed, wrap up, hand off, task complete, finishing work, workspace done, afronden, afgerond, inleveren, cleanup, test de fix, check result, prove it works, check of het klopt, validate, valideren, werkt het, is het klaar.
ui-ux
Use when the request is to design, redesign, polish, review, or implement UI/UX for web or mobile interfaces, landing pages, dashboards, flows, or visual systems. Common triggers: design a ui, redesign this page, improve ux, polish the frontend, landing page design, dashboard design, mobile app ui, design system, ui review, redesign the frontend, improve this page, landing page, mobile ui, color palette, typography, font pairing, ui style, accessibility, animation design, data visualization, ux patterns, design system generation.
write-skill
Use when creating or revising skills, or when agent behavior reveals a reusable workflow improvement, routing gap, or missing guardrail that should become a skill, plugin, or prompt-pack follow-up. Common triggers: create skill, revise skill, skill design, trigger-focused, write skills, improve skills, skill improvement, skill gap, workflow improvement, routing gap, missing guardrail, prompt pack improvement, reusable improvement, agent missed, auto improvement.
github-issues
Use when creating a GitHub issue from a bug report, review finding, failed attempt, or requested follow-up, especially when the repo may need to be inferred from the current checkout. Common triggers: create issue, github issue, file issue, gh issue create, bug report.
gh-inbox
Process the current repository's GitHub issues and discussions, triage activity, reply when clear, and persist inbox state. Use when asked to check the GitHub inbox, triage issues or discussions, or catch up on repo activity. Common triggers: github inbox, gh inbox, triage issues, check issues, check discussions, reply to issue, process inbox, gh-inbox.
design-review
Review an existing design, UI, or copy for AI-generated default patterns and quality issues before shipping or handoff. Common triggers: review this design, check for AI slop, does this look AI-generated, design review, audit the design, does this look premium, review the frontend, check this UI.
spec
Use when building a requirements spec or design brief before code — formalizing capture, decisions, and validation gates for non-trivial or enterprise-style work. Traceability + human-owned validation before handover. Common triggers: spec, specify requirements, requirements spec, requirements capture, design brief, decision register, requirements traceability, traceable requirements, validation gate, readiness gate, handover package, spec before build, truth spine, requirements-driven, requirements engineering, formalize requirements, requirements specification, engineering contract, invariant, proof obligation, counterexample, compatibility-sensitive, material requirement.
text-writing
Produce human-sounding text that avoids detectable AI writing patterns. Use for tweets, emails, articles, bios, captions, reports, copy, messages, LinkedIn posts, cover letters, README files — any output that must not read as AI-generated. Common triggers: anti-slop, make this sound human, sound human, not AI, does not read like AI, write a tweet, draft email, write an email, cover letter, linkedin post, newsletter, blog post, copywriting, schrijf als mens, niet AI, menselijk laten klinken.
deep-research
Use when a complex technical question needs autonomous, multi-source investigation, contradiction analysis, and an actionable cited handoff. Common triggers: deep research, exhaustive research, comprehensive investigation, complex technical investigation, complex contested high-stakes question, complex compatibility question, complex compatibility issue, complex question, contested research, high-stakes research, full compatibility investigation, compare competing implementations, compare local and upstream implementations, investigate historical changes, determine protocol behaviour, investigate protocol behavior, protocol behavior exhaustively, multiple sources, research everything relevant, investigate open issues, open issues comprehensively, compare against upstream.
design
Use when the request is to design, redesign, polish, review, or implement UI/UX for web or mobile interfaces, landing pages, dashboards, flows, or visual systems. Common triggers: design a ui, redesign this page, improve ux, polish the frontend, landing page design, dashboard design, mobile app ui, design system, ui review, redesign the frontend, improve this page, landing page, mobile ui, color palette, typography, font pairing, ui style, accessibility, animation design, data visualization, ux patterns, design system generation.
research
Use when a question needs bounded fact-finding across repository, documentation, code, or external sources before an answer or decision. Common triggers: research this, research question, find evidence, compare sources, investigate current state, look into this technology, research documentation.
kaizen
Default baseline skill for normal software work: small safe iterations, built-in validation, no unnecessary pauses between clear next steps. Includes mode selection (direct, batch, delegate) and cheap-first escalation for bounded mechanical chores. Common triggers: kaizen, autopilot, keep going, continue without waiting, do not stop, don't stop, ga door, werk door, volgende logische stap, volgende stap, start working, start coding, get started, zonder te wachten, niet stoppen, gewoon doorgaan, ga verder, implement this, fix this, add this, pas dit aan, maak dit af, implement, implementeer dit, keep coding, continue implementation, work through steps, code change, start implementing, start implementation, batch edits, delegate work, subagent, maak dit werkend.
writing-nebu-skills
Use when creating or revising skills, or when agent behavior reveals a reusable workflow improvement, routing gap, or missing guardrail that should become a skill, plugin, or prompt-pack follow-up. Common triggers: create skill, revise skill, skill design, trigger-focused, write skills, improve skills, skill improvement, skill gap, workflow improvement, routing gap, missing guardrail, prompt pack improvement, reusable improvement, agent missed, auto improvement.
observability
Use when instrumenting a feature for production — adding or improving logging, metrics, tracing, or alerting — or when production issues are reported and the current telemetry cannot explain them. Common triggers: observability, instrumentation, add logging, structured logging, log levels, add metrics, adding metrics, metrics dashboard, set up metrics, set up tracing, distributed tracing, opentelemetry, set up alerting, alerting on, alert rule, runbook, telemetry setup, app telemetry, monitor this feature, how do we observe, what is working in production, monitoring alerts, instrument this, production visibility.
kickoff
Use when the goal, constraints, or success criteria are not yet crisp — from fuzzy design ideas through ambiguous scope to multi-file planning. Covers brainstorming, scoping, and execution planning in one skill. Common triggers: brainstorm, brainstormen, fuzzy idea, design tradeoff, unsure what to build, product direction, idee uitwerken, ambiguous request, unclear scope, behavior-changing work, fuzzy requirements, what should we build, wat moeten we bouwen, wat moeten we maken, best approach, how should we approach this, not sure where to start, start by clarifying, start with questions, ik weet niet waar te beginnen, hoe pakken we dit aan, plan, plannen, multi-file work, multi-phase work, migration, sequencing risk, start planning, start with a plan, werk voorplannen.
improve
Use when the codebase needs a structured audit, audit-driven plans, execution of those plans, or a focused refactoring/simplification pass. Covers correctness, security, performance, tech debt, migrations, DX, direction, and code cleanup in one skill. Common triggers: improve, audit, tech debt, tech debt audit, audit codebase, improve codebase, direction, audit and plan, refactor this, refactoren, code cleanup, opschonen, simplify this code, vereenvoudigen, remove over-engineering, deduplicate logic, restructure this code, reduce complexity, untangle this, clean architecture mess.
ui-ux
Use when the request is to design, redesign, polish, review, or implement UI/UX for web or mobile interfaces, landing pages, dashboards, flows, or visual systems. Common triggers: design a ui, redesign this page, improve ux, polish the frontend, landing page design, dashboard design, mobile app ui, design system, ui review, redesign the frontend, improve this page, landing page, mobile ui, color palette, typography, font pairing, ui style, accessibility, animation design, data visualization, ux patterns, design system generation.
write-skill
Use when creating or revising skills, or when agent behavior reveals a reusable workflow improvement, routing gap, or missing guardrail that should become a skill, plugin, or prompt-pack follow-up. Common triggers: create skill, revise skill, skill design, trigger-focused, write skills, improve skills, skill improvement, skill gap, workflow improvement, routing gap, missing guardrail, prompt pack improvement, reusable improvement, agent missed, auto improvement.
code-review
Use when code changed and a meaningful diff is ready; fresh eyes should catch requirement gaps, regressions, or risky design mistakes before handoff or success claims. Common triggers: review, nakijken, pull request, code review, fresh eyes, start reviewing, review deze wijziging, check de wijziging, review changes, second look, bekijk de diff, controleer de code, code check, diff review, PR review.
debugging
Use when a bug, failing test, or broken build is not already explained by a clear local mistake, or when a first fix did not work. Common triggers: bug, failing test, broken build, debug, debuggen, error, crash, stack trace, not working, does not work, broke, start debugging, start investigating, fout opsporen, slow startup, timeout, hangt, hanging, crash loop, None, target_temp, niet werkend, doet het niet, malfunction, storing.
session-review
Use after completing work to reflect on skill usage, identify gaps, and file improvement issues in the nebu-skills repo. Also handles general GitHub issue creation from bug reports, review findings, and follow-ups. Common triggers: retrospective, retro, reflect on session, how did i use skills, file an issue, create issue, github issue, file issue, gh issue create.
verification
Use when about to claim something works, is fixed, or ready to hand off — and when a task changed one or more repos and needs intentional cleanup before stopping. Common triggers: verify, verifiëren, prove, controleren of het werkt, bewijzen dat het werkt, claim success, done, finished, ready, handoff, klaar, gereed, wrap up, hand off, task complete, finishing work, workspace done, afronden, afgerond, inleveren, cleanup, test de fix, check result, prove it works, check of het klopt, validate, valideren, werkt het, is het klaar.
agent-workflows
Use when coordinating multi-agent work, parallel execution, task handoff, shared context, or clean session shutdown across multiple agents or terminals. Especially useful when the host supports subagents, hooks, or shared context. Common triggers: multi-agent, parallel work, agent coordination, task handoff, subagent delegation, version bump, bump version, release notes, changelog, tag release, release prep.
develop
Default baseline skill for normal software work: small safe iterations, built-in validation, no unnecessary pauses between clear next steps. Includes mode selection (direct, batch, delegate) and cheap-first escalation for bounded mechanical chores. Common triggers: develop, kaizen, autopilot, keep going, continue without waiting, do not stop, don't stop, ga door, werk door, volgende logische stap, volgende stap, start working, start coding, get started, zonder te wachten, niet stoppen, gewoon doorgaan, ga verder, implement this, fix this, add this, pas dit aan, maak dit af, implement, implementeer dit, keep coding, continue implementation, work through steps, code change, start implementing, start implementation, batch edits, delegate work, subagent, maak dit werkend, rewrite, herbouw, coordinator, coördinator, omwerken, entity rewrite, refactor, code aanpassen, werk dit bij, build this, development.
intake
Use when the goal, constraints, or success criteria are not yet crisp — from fuzzy design ideas through ambiguous scope to multi-file planning. Covers brainstorming, scoping, and execution planning in one skill. Common triggers: brainstorm, brainstormen, fuzzy idea, design tradeoff, unsure what to build, product direction, idee uitwerken, ambiguous request, unclear scope, behavior-changing work, fuzzy requirements, what should we build, wat moeten we bouwen, wat moeten we maken, best approach, how should we approach this, not sure where to start, start by clarifying, start with questions, ik weet niet waar te beginnen, hoe pakken we dit aan, plan, plannen, multi-file work, multi-phase work, migration, sequencing risk, staged refactor, stages, service by service, per service, dependency chain, sequential steps, per laag, stap voor stap, start planning, start with a plan, werk voorplannen, we moeten dit aanpakken, laten we dit doen, we moeten, laten we, pair programming, samenwerken,...
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.