All categories
AI & Agents
LLMs, agent workflows, RAG, MCP, prompting, and AI app patterns
- 307,719
- 12,822
Security grades appear on each card once the skill has been scanned. Newly imported skills may briefly show without a grade until the backfill job runs.
Open in full browserBrowse ai & agents skills
Showing 3,169–3,192 of 307,719 skills
- Crisis ModeOperating under fire: production incidents, deadline collapses, demo-eve disasters. Stabilize, diagnose, fix, learn — in that order, with a cool head. Trigger: load when something is broken NOW and the cost of downtime is running — outages, data issues, launch-day failures, demo-tomorrow panics.Votes: 0GitHub stars: 2
- React NextDecision criteria for React/Next.js feature work: server/client split, state placement, component boundaries, data flow. Trigger: load when building or restructuring React components, pages, or data fetching in a React app. The server/client and data-fetching patterns apply to Next.js App Router (or a framework with server rendering and loaders); a Vite SPA or React Native takes the state ladder and the boundaries and skips those. Complements (does not replace) any `react-19` / `nextjs-*` ski...Votes: 0GitHub stars: 2
- PairingHow to work as a peer: reviewing a teammate's code, debugging together, and splitting work. Trigger: load when reviewing the user's code as a colleague, co-debugging an issue, or dividing a task between you and the user.Votes: 0GitHub stars: 2
- Fullstack BoundariesJudgment for work that crosses the wire, in any stack: API contracts, runtime validation at edges, where logic lives, errors end-to-end. Trigger: load when work crosses the wire — designing/consuming APIs, handling forms, touching the data layer, or wiring frontend to backend.Votes: 0GitHub stars: 2
- Code HealthClean code judgment and technical debt management: what to improve while touching code, when to take debt consciously, and which debt to pay first. Trigger: load when deciding whether an improvement belongs in this diff or in a ticket, when tempted to take a shortcut, or when choosing which debt to pay first. The clean-code baseline is always on and needs no skill.Votes: 0GitHub stars: 2
- Threat ModelingMapping attack surface and trust boundaries before reading code: entry points, data flows, trust zones, and adversary goals. Trigger: load when starting a security assessment, designing a feature securely, or asked "what could an attacker do here / where are we exposed?".Votes: 0GitHub stars: 2
- RemediationTurning security findings into fixes: the actionable report, severity ranking, fix-the-class prescriptions, and verification. Trigger: load when reporting vulnerabilities, prioritizing a findings list, prescribing fixes, or verifying a security fix.Votes: 0GitHub stars: 2
- Code AuditExhaustive code auditing for vulnerability classes: injection, authz, auth, secrets, crypto, deps, and the method for reading code adversarially. Trigger: load when auditing code for security flaws, reviewing a diff for vulnerabilities, or hunting a specific vuln class.Votes: 0GitHub stars: 2
- Test DesignJudgment for writing and reviewing test code: what to test at which level, how to keep tests trustworthy, and how to audit an existing suite. Trigger: load when writing test code, reviewing a test suite, or deciding what automated coverage a feature needs.Votes: 0GitHub stars: 2
- Flow HuntingSystematic hunting of unconsidered paths: edge cases, interruptions, state collisions, and hostile inputs. Trigger: load when exploring a feature or flow for potential bugs, reviewing a spec for gaps, or asked "what could break here?".Votes: 0GitHub stars: 2
- Bug ReportingTurning findings into action: minimal reproductions, severity honesty, and the collaboration loop with dev and ux-ui. Trigger: load when reporting a found issue, triaging findings, verifying a fix, or deciding whether something is a bug, a design gap, or an improvement.Votes: 0GitHub stars: 2
- StakeholdersAlignment and communication: managing stakeholders, communicating product decisions, saying no gracefully, and setting expectations. Trigger: load when aligning stakeholders, communicating a decision or roadmap, handling a pushed-in request, or managing expectations up and across.Votes: 0GitHub stars: 2
- DiscoveryValidating that a problem is real and worth solving before anyone builds: user research, assumption-testing, and opportunity assessment. Trigger: load when evaluating a new feature/idea, deciding whether to build something, or when a request arrives as a solution and the problem is unclear.Votes: 0GitHub stars: 2
- BacklogTurning validated problems into buildable work: user stories, acceptance criteria, prioritization, and roadmap. Trigger: load when writing requirements/stories, prioritizing a backlog, planning a roadmap, or defining what "done" means for a product change.Votes: 0GitHub stars: 2
- VerificationHabits for proving claims before making them: the evidence ladder, when to verify, and how to phrase confidence honestly. Trigger: load before claiming done on work with a runtime surface, before building on a technical assertion you have not checked, and before fixing a bug you have not reproduced. Not needed for a one-line factual answer — the base context covers that.Votes: 0GitHub stars: 2
- Next StepDecision model for choosing what to do next: continue, re-plan, stop, or escalate to the user. Trigger: load after a step fails or surprises you, before ending a turn on a task that took more than one step, before starting a task of more than a handful of steps, or when resuming one ("seguimos con lo de ayer"). Not needed when the next action is obvious and reversible.Votes: 0GitHub stars: 2
- DecompositionHow to break non-trivial work into independently verifiable steps ordered by uncertainty reduction. Trigger: load before planning any task that needs more than one action, spans multiple files, or contains an assumption you haven't confirmed.Votes: 0GitHub stars: 2
- Game MechanicsThe gamification toolbox with judgment: core loops, progression, rewards, social mechanics — what each one does, whom it serves, and its shadow side. Trigger: load when choosing, designing, or tuning game mechanics for a product flow (only after flow-analysis has run).Votes: 0GitHub stars: 2
- Flow AnalysisHow to walk a product flow as an analyst: map it from the user's side, audit friction and motivation step by step, and simplify before gamifying. Trigger: load when asked to analyze, review, or improve any user flow, funnel, or journey — always BEFORE proposing mechanics.Votes: 0GitHub stars: 2
- EngagementAttracting users and bringing them back: onboarding as a game's first session, habit loops, return triggers, and honest engagement metrics. Trigger: load when working on onboarding, activation, retention, churn, or when asked "how do we get users to come back / stick around".Votes: 0GitHub stars: 2
- Team HealthKeeping the team (of agents and people) effective and sustainable: minimal process, retrospectives, unblocking, and continuous improvement. Trigger: load when designing/adjusting process, running a retro, addressing recurring friction, or when the team's way of working needs tuning.Votes: 0GitHub stars: 2
- OrchestrationRouting work across the agent team and composing multiple agents on shared tasks: the roster map, routing heuristics, and handoff choreography. Trigger: load when a task spans multiple specialties, when unsure which agent fits, or when coordinating several agents on one piece of work.Votes: 0GitHub stars: 2
- DeliveryGetting work delivered predictably: planning, estimation, scoping, tracking, and de-risking. Trigger: load when planning work, estimating, breaking down an epic, tracking progress, or a project is slipping/at risk.Votes: 0GitHub stars: 2
- ObservabilityKnowing what production is doing: logs, metrics, traces, meaningful alerts, and SLOs. Trigger: load when adding monitoring, designing alerts, debugging a production issue blind, or defining what "healthy" means for a service.Votes: 0GitHub stars: 2