zap-coding-agent
- 62 skills
- 0 followers
- 10 hours ago last updated
- ▌ Karpathy Neural Nets · zap-coding-agentUse when building or debugging neural networks, deciding how to structure ML code for understanding, training a model that isn't converging, or approaching any complex technical system using Karpathy's "understand every line" pedagogy. Covers his recipe for training neural nets, Software 2.0 philosophy, and his build-from-scratch approach to deep understanding.
- ▌ Karpathy Vibe Coding · zap-coding-agentUse when working with AI coding assistants (Claude, Copilot, Cursor, etc.) as a core part of development — deciding what to delegate to AI vs write yourself, maintaining code understanding while using AI generation, catching AI mistakes, and applying Karpathy's philosophy of "vibe coding" for the right tasks and manual coding for the ones where understanding matters.
- ▌ Pocock Write A Skill · zap-coding-agentUse when designing and writing a new agent skill — applies Matt Pocock's discipline around description-as-trigger, progressive disclosure, and bundling reference material. Triggers on "write a skill", "create a skill", or when reviewing whether an existing skill's description or structure is correct.
- ▌ Splunk Spl Authoring · zap-coding-agent bundleUse when writing, fixing, or reviewing Splunk SPL searches — building a query from scratch, extracting fields, aggregating with stats/timechart/tstats, parsing raw logs, or when a search returns no/wrong results or runs slowly. Covers the search pipeline, command order, and idiomatic SPL.
- ▌ Splunk Threat Hunting · zap-coding-agentUse when proactively hunting for threats in Splunk that existing alerts didn't catch — forming a hypothesis, exploring data for anomalies, looking for lateral movement, beaconing, living-off-the-land, persistence, or suspicious patterns, and turning findings into detections. Distinct from incident investigation (no known alert) and from writing detections (hunting discovers what to detect).
- ▌ Developer Productivity · zap-coding-agentUse when improving developer workflow, reducing friction in a codebase, designing local development environments, making CI faster, eliminating toil, evaluating developer experience, or thinking about how team practices (async communication, PR size, feedback loops) affect throughput. Covers individual and team-level productivity.
- ▌ Persistence Repository · zap-coding-agentUse when designing or reviewing the persistence layer — repository interfaces and implementations, the repository vs DAO distinction, mapping domain aggregates to/from ORM entities or rows, query placement, and keeping the database out of the domain. Catches ORM entities leaking as domain models and business logic in repositories. Technology-agnostic.
- ▌ Karpathy LLM Intuitions · zap-coding-agentUse when building intuition for how LLMs work — attention mechanisms, tokenization gotchas, why transformers work, context window behavior, temperature/sampling, prompt engineering from first principles, or when debugging why a model gives unexpected outputs. Applies Karpathy's "understand the math, not just the API" approach.
- ▌ Splunk Spl Optimization · zap-coding-agentUse when a Splunk search is slow, times out, hits subsearch/memory limits, scans too many events, or costs too much — or when reviewing SPL for performance before scheduling it as an alert/dashboard panel. Covers filtering early, tstats/data models, avoiding join/transaction, and search-cost diagnosis.
- ▌ Loop Stopping Conditions · zap-coding-agentUse when a loop doesn't terminate, runs forever on a bad goal, burns budget without making progress, or needs a reliable halt strategy. Covers primary stopping conditions (goal met), fallback budgets (max turns/cost/no-progress), and escalation when neither fires.
- ▌ Splunk Cim Normalization · zap-coding-agentUse when normalizing Splunk fields to standard names so searches, dashboards, and detections work across data sources — mapping vendor fields to CIM (Common Information Model), building/validating data models, tagging events, or when a correlation search or Enterprise Security panel returns nothing because fields don't match. Covers CIM data models, field aliases, tags, eventtypes, and acceleration.
- ▌ Splunk Log Investigation · zap-coding-agentUse when investigating an incident, outage, error spike, latency regression, or anomaly using Splunk logs — going from a symptom ("500s started at 2pm", "checkout is slow", "users can't log in") to root cause. Covers scoping, correlating across sources, tracing a request, and building the timeline.
- ▌ Fowler Enterprise Patterns · zap-coding-agentUse when designing data access patterns — choosing between Active Record, Data Mapper, Repository, Table Module, and Service Layer; understanding transaction scripts vs domain models; mapping objects to databases; or reviewing code where persistence and domain logic are tangled. From Fowler's Patterns of Enterprise Application Architecture (PoEAA).
- ▌ Pocock Typescript Generics · zap-coding-agentUse when TypeScript generics feel too hard or too loose — writing generic utilities, using conditional types with infer, mapping over types, building type-safe function overloads, or when you get "Type instantiation is excessively deep" errors. Applies Pocock's advanced type-level programming approach.
- ▌ Splunk Detection Engineering · zap-coding-agent bundleUse when building, tuning, or reviewing Splunk security detections — correlation searches, scheduled alerts, Enterprise Security notable events, risk-based alerting (RBA), or mapping detections to MITRE ATT&CK. Also when a detection is too noisy, misses true positives, or you need to reduce false positives. Covers detection logic, thresholds, throttling, and the detection lifecycle.
- ▌ Pocock Tsconfig · zap-coding-agentUse when configuring TypeScript — setting up tsconfig.json for a new project, understanding what strict mode enables, configuring path aliases, setting up project references for monorepos, debugging "cannot find module" or declaration file errors, or understanding which tsconfig options actually matter. Applies Pocock's (Total TypeScript) tsconfig guidance.
- ▌ React Internals · zap-coding-agentUse when you need to understand *why* React works the way it does — Fiber reconciliation, the rendering model, Concurrent Mode, hooks internals, or when debugging unexpected re-renders, stale closures, batching behavior, or tearing. Understanding the internals predicts behavior without memorizing rules.
- ▌ Caveman Compress · zap-coding-agentUse when a CLAUDE.md, memory file, or large instruction document is bloated with filler and needs to be compressed to save input tokens — rewrites it in caveman style while preserving all technical substance. Triggers on /caveman-compress FILEPATH or "compress my CLAUDE.md".
- ▌ Hickey Simplicity · zap-coding-agentUse when evaluating whether a design is genuinely simple or just familiar — distinguishing simple from easy, identifying complected concepts that should be separated, deciding whether to add an abstraction, or reviewing architecture for hidden complexity. Applies Rich Hickey's "Simple Made Easy" philosophy.
- ▌ Martin Clean Arch · zap-coding-agentUse when designing the high-level structure of a system — applying Uncle Bob's Clean Architecture (the concentric rings), understanding the Dependency Rule, designing use cases as the architectural center, separating policy from mechanism, or when reviewing whether a codebase's architecture protects the business rules from framework/database/UI changes.
- ▌ Nextjs App Router · zap-coding-agentUse when building with Next.js 13+/14+ App Router — understanding Server Components vs Client Components, data fetching with async/await or fetch(), caching model (static/dynamic/revalidation), Server Actions, route handlers, streaming with Suspense, metadata, and when something renders on the server vs client. Covers the mental model shift from Pages Router.
- ▌
- ▌
- ▌ Splunk Dashboards · zap-coding-agentUse when building or improving Splunk dashboards, reports, or visualizations — Dashboard Studio or Simple XML, choosing the right chart, adding inputs/filters/drilldowns, defining KPIs, making panels load fast with base searches or accelerated data models, or fixing dashboards that are slow, cluttered, or hard to read.
- ▌ Clean Architecture · zap-coding-agentUse when deciding which layer code belongs in, reviewing whether a controller/service/domain/repository is doing too much, untangling a "fat controller" or "anemic/God service", or enforcing the dependency rule between API, application, domain, and persistence layers. Technology-agnostic; the glue skill for the layered-* suite.
- ▌
- ▌ Loop Maker Checker · zap-coding-agentUse when an agent loop needs a quality gate, an agent is self-certifying its own output, outputs are unreliable without independent review, or you need to split code generation from code verification inside a loop. Practical implementation of the maker-checker pattern.
- ▌ Karpathy Software 2 · zap-coding-agentUse when deciding whether a problem should be solved with traditional code vs a neural network, when framing ML as an engineering discipline (not research), understanding dataset curation as software engineering, or evaluating the "is this worth MLing?" question. Applies Karpathy's Software 2.0 thesis.
- ▌ Cherny Maker Checker · zap-coding-agentUse when designing a multi-agent system, an agent is self-certifying its own work as complete, outputs aren't trustworthy without independent review, or a loop needs a quality gate. Applies Boris Cherny's maker-checker pattern — the most important structural move inside any agent loop.
- ▌ Fowler Microservices · zap-coding-agentUse when deciding whether to use microservices, how to decompose a monolith, setting service boundaries, designing inter-service communication, or when a microservices system is causing more pain than it's solving. Applies Martin Fowler's nuanced take — he coined the term but also wrote "MonolithFirst".
- ▌ Hickey Data Oriented · zap-coding-agentUse when deciding between data-oriented and object-oriented design, building systems that process or transform data across boundaries, designing APIs that are easy to combine and extend, evaluating whether encapsulation is helping or hurting, or when a system is rigid and hard to extend. Applies Rich Hickey's data-oriented programming philosophy.
- ▌ Caveman · zap-coding-agentUse when token efficiency matters, responses are too verbose, or user says "caveman mode", "talk like caveman", "be brief", "less tokens", or invokes /caveman. Switches to ultra-compressed communication that cuts ~65% output tokens while keeping full technical accuracy.
- ▌ Handoff · zap-coding-agentUse when a long conversation needs to be passed to a fresh agent session — compacts the current context into a handoff document another agent can pick up without losing thread. Triggers on "handoff", "summarise for next session", or "prepare context for new agent".
- ▌ Beck Tdd · zap-coding-agentUse when implementing any feature, fixing any bug, or when deciding whether tests are "good enough" — applying Kent Beck's Test-Driven Development as its originator practiced it. Covers Red/Green/Refactor, the simple design rules, YAGNI, baby steps, and the key forcing questions Beck uses to keep code honest.
- ▌ Grill Me · zap-coding-agentUse when the user wants to stress-test a plan or design before writing code — asks probing questions one at a time, walks every branch of the decision tree, and surfaces implicit assumptions. Triggers on "grill me", "stress-test this plan", or "challenge my design".
- ▌ 100k Linux · zap-coding-agentUse when contributing to kernel-style projects, writing patch-based submissions, doing code review in a direct/high-signal culture, understanding Linus Torvalds' engineering philosophy, or designing systems where correctness and simplicity are absolute constraints. Covers the Linux development model, patch culture, and Linus's core engineering principles.
- ▌ 100k React · zap-coding-agentUse when making architectural decisions for a React-based project, contributing to React-style open source, understanding why React's API is shaped the way it is, or navigating the React RFC process. Covers the React team's decision-making philosophy, backward-compatibility constraints, the "learn once write anywhere" principle, and how they evaluate proposals. Distinct from react-internals (which covers the Fiber technical model).
- ▌ Tan Design · zap-coding-agentUse when making design decisions for a product, evaluating whether a UI is good, working with designers as an engineer, applying design thinking to technical problems, or understanding how Garry Tan's design background shapes his product philosophy. Covers taste, visual hierarchy, the engineer-designer collaboration, and how good design compounds.
- ▌ Cherny Loop · zap-coding-agentUse when designing an agentic system, moving beyond single prompts to autonomous agent loops, deciding whether to build a loop or write a prompt, or when work should run unattended on a schedule. Applies Boris Cherny's loop engineering framework — the paradigm shift from prompt-operator to loop-architect.
- ▌ Dto Mapping · zap-coding-agentUse when designing DTOs / request-response models / commands, or mapping between them and domain objects — at the API edge or any boundary. Catches leaking ORM/domain entities to clients, over-posting/mass-assignment, behavior creeping into DTOs, and mapping logic scattered through controllers/services. Technology-agnostic.
- ▌ Loop Design · zap-coding-agentUse when building an agent loop from scratch — designing the trigger, goal, actions, verification, and memory of an autonomous agent system. Applies when a recurring task should run unattended, when a single agent prompt isn't enough, or when you need compounding leverage from a system you build once.
- ▌ Ml Engineer · zap-coding-agentUse when approaching problems as an ML engineer — framing ML problems correctly, choosing between ML vs non-ML solutions, designing training/evaluation pipelines, preventing data leakage, setting up experiment tracking, deploying models, and monitoring model quality in production. Covers the ML engineering mindset alongside the software engineering practices that make ML reliable.
- ▌ QA Engineer · zap-coding-agentUse when approaching software quality as a QA engineer — building a testing strategy, deciding what to test and at which layer, writing effective test cases, shifting quality left, reviewing code for testability, designing CI quality gates, or thinking about exploratory vs automated testing. Covers QA mindset, test pyramid, and quality advocacy.
- ▌ Tan Product · zap-coding-agentUse when making product decisions, evaluating startup ideas, deciding whether to build something, prioritizing features, designing for early users, or thinking about the relationship between engineering and product strategy. Applies Garry Tan's (YC president, ex-Initialized Capital, ex-Posterous) product and founder philosophy — founder mode, talking to users, design-engineering integration.
- ▌ Tan Startup · zap-coding-agentUse when making startup strategy decisions — evaluating product-market fit signals, deciding when to pivot vs persist, thinking about growth, fundraising readiness, early team building, or applying YC's frameworks to company decisions. Applies Garry Tan's (YC president) frameworks for company building.
- ▌ Domain Model · zap-coding-agentUse when designing or reviewing the domain layer — entities, value objects, aggregates, invariants, domain logic and domain events. Catches the "anemic domain model" (logic-less data bags with all rules in services) and domain code that leaks framework/ORM/persistence concerns. Technology-agnostic; DDD building blocks.
- ▌ Martin Solid · zap-coding-agentUse when reviewing class/module design, deciding whether to extract/split/combine code, evaluating coupling and cohesion, or writing a justification for a design decision. Applies Robert C. Martin's SOLID principles with the original formulations, failure modes each prevents, and the common over-application traps.
- ▌ Pocock Teach · zap-coding-agentUse when the user wants to learn something deeply over multiple sessions — applies Matt Pocock's mission-first, stateful teaching system with lessons, reference docs, and learning records. Triggers on "teach me", "I want to learn X properly", or when learning needs to persist across sessions.
- ▌ Service Layer · zap-coding-agentUse when writing or reviewing an application/service/use-case layer — orchestrating a business operation across domain objects and repositories, setting transaction boundaries, publishing events, and coordinating side effects. Catches "God services" stuffed with business rules, mapping, or framework/SQL code. Technology-agnostic.
- ▌ Write A Skill · zap-coding-agentUse when the user wants to create a new agent skill — gathers requirements, drafts SKILL.md with proper description format, adds reference files if needed, and reviews with the user. Triggers on "create a skill", "write a skill", "build a new skill", or "add a skill".
- ▌ API Controller · zap-coding-agentUse when writing or reviewing an API/controller/handler layer (REST, GraphQL resolver, gRPC, CLI command, message consumer) — request deserialization, input validation, auth checks, status codes, error mapping, and keeping controllers thin. Catches "fat controllers" with business logic or direct DB access. Technology-agnostic.
- ▌ Caveman Commit · zap-coding-agentUse when writing a git commit message — generates terse Conventional Commits format with subject ≤50 chars, body only when the "why" isn't obvious. Triggers on "write a commit", "commit message", "generate commit", or /caveman-commit.
- ▌
- ▌ Pocock Handoff · zap-coding-agentUse when a long conversation needs to be handed off to a fresh agent — applies Matt Pocock's discipline of clean context transfer so the next session starts with full understanding rather than re-deriving everything. Triggers on "handoff", "prepare for next session", or "summarise for a new agent".
- ▌ React Patterns · zap-coding-agentUse when designing React components — choosing between hooks, compound components, render props, and composition patterns; avoiding prop drilling; managing shared state; or when a component is growing too large and needs splitting. Covers idiomatic React composition patterns.
- ▌ 100k Kubernetes · zap-coding-agentUse when designing distributed system APIs, control planes, or systems that need eventual consistency — applying Kubernetes' declarative API design, controller/reconciliation loop pattern, operator pattern, or level-triggered vs edge-triggered design. Also when evaluating whether a system should be imperative or declarative.
- ▌ 100k Typescript · zap-coding-agentUse when designing TypeScript type signatures, choosing between generics/unions/intersections/conditional types, deciding how much type coverage to add, debugging "type instantiation too deep" or unsound types, or understanding why TypeScript makes specific trade-offs (structural typing, soundness vs completeness). Covers the TypeScript team's design goals and the practical implications.
- ▌ Devops Engineer · zap-coding-agentUse when thinking about deployment pipelines, infrastructure-as-code, observability, reliability engineering, incident response, or the DevOps/SRE mindset — CI/CD design, toil reduction, SLOs, on-call, runbooks, blast radius reduction, and the philosophy of treating operations as a software engineering problem. Technology-agnostic.
- ▌ Nextjs Patterns · zap-coding-agentUse when architecting a Next.js App Router application — organizing routes and layouts, handling auth/middleware, structuring Server vs Client boundaries, setting up metadata and SEO, managing environment variables, monorepo patterns, and error/loading/not-found conventions. Covers structural decisions and file-system conventions.
- ▌ Pocock Grill Me · zap-coding-agentUse when the user wants to stress-test a plan or design before writing code — applies Matt Pocock's discipline of resolving every design branch before touching the keyboard. Triggers on "grill me", "stress-test this", or when a plan has unresolved forks that will become bugs.
- ▌ Teach · zap-coding-agentUse when the user wants to learn a new skill or concept over multiple sessions — sets up a stateful teaching workspace with missions, lessons, and reference documents. Triggers on "teach me", "I want to learn", or "explain X from scratch".
- ▌ Beck Xp · zap-coding-agentUse when structuring team engineering practices — pair programming, collective code ownership, sustainable pace, continuous integration (the original meaning), small releases, planning game, or when a team is moving slowly because of process overhead or quality problems. Applies Kent Beck's Extreme Programming practices beyond TDD.