nimadorostkar

agent-instructions

Use when writing project instructions for a coding agent (CLAUDE.md, AGENTS.md, or equivalent). Covers what belongs in them, what does not, structure, and keeping them accurate as the project changes.

agent-memory

Use when an agent needs state that survives a session or a context compaction. Covers what to persist, file-based memory, structuring notes for retrieval, and preventing memory from becoming stale.

hooks

Use when automating agent behavior with lifecycle hooks. Covers hook events, deterministic enforcement of rules the model should not be trusted to remember, and avoiding hooks that make an agent unusable.

plugin-development

Use when packaging skills, commands, hooks, and MCP servers into a distributable plugin. Covers manifest structure, bundling, versioning, testing, and distribution through a marketplace.

skill-authoring

Use when writing a new skill for an AI agent. Covers scoping, description writing for reliable triggering, progressive disclosure, and the difference between a skill and a prompt.

skill-review

Use when reviewing or improving an existing agent skill. Covers triggering accuracy, content quality, redundancy with the base model, and measuring whether the skill actually changes behavior.

slash-commands

Use when creating reusable slash commands for an agent. Covers when a command beats a skill, argument handling, composing tool calls, and writing commands that are worth typing.

subagents

Use when delegating work to a separate agent with its own context. Covers when delegation pays off, designing a subagent's scope and prompt, parallel fan-out, and the cost of a cold start.

agent-design

Use when building an LLM agent that uses tools over multiple steps. Covers tool design, the agent loop, error recovery, termination, human checkpoints, and knowing when an agent is the wrong architecture.

context-engineering

Use when managing what an LLM sees. Covers context-window budgeting, retrieval and compaction, memory across turns, tool-result pruning, and the failure modes that come from too much context rather than too little.

fine-tuning

Use when considering fine-tuning a model. Covers when fine-tuning beats prompting or RAG, dataset construction, LoRA and full fine-tuning, evaluation, and the failure modes that waste the effort.

llm-evaluation

Use when measuring the quality of an LLM feature. Covers building an evaluation set, choosing metrics, LLM-as-judge and its pitfalls, regression testing prompts, and evaluating in production.

llm-integration

Use when integrating an LLM API into an application. Covers streaming, retries and rate limits, timeouts, caching, fallback across providers, and the production concerns that a tutorial integration ignores.

mcp-server

Use when building a Model Context Protocol server. Covers tool, resource, and prompt design, transport choice, authentication, error handling, and testing an MCP server against a real client.

ml-pipeline

Use when building or operating a machine learning pipeline. Covers feature engineering, training reproducibility, train/serve skew, deployment, monitoring for drift, and retraining.

model-selection

Use when choosing which LLM to use for a task. Covers matching capability to task, cost and latency trade-offs, routing between models, benchmark skepticism, and evaluating on your own data.

prompt-engineering

Use when writing or improving prompts for a language model. Covers instruction structure, examples, reasoning elicitation, output formatting, and systematically diagnosing why a prompt fails.

rag

Use when building retrieval-augmented generation. Covers chunking, embedding and hybrid search, reranking, grounding and citation, and diagnosing whether a bad answer is a retrieval failure or a generation failure.

structured-output

Use when an LLM must return machine-readable data. Covers schema design for models, native structured-output modes, validation and repair, and extraction that survives contact with messy input.

api-design

Use when designing or reviewing an HTTP API. Covers resource modeling, status codes, pagination, idempotency, versioning, error formats, and the contract details that break clients when you get them wrong.

background-jobs

Use when designing asynchronous job processing. Covers queue selection, idempotency, retry and backoff policy, scheduling, poison messages, and observability for work that happens outside the request.

caching

Use when adding or debugging a cache. Covers cache placement, invalidation strategies, stampede protection, TTL selection, and the consistency you are trading away.

django

Use when building Django applications. Covers ORM query performance, model design, migrations, Django REST Framework, security defaults, and testing.

event-driven-architecture

Use when designing systems around events and message queues. Covers event schema design, delivery guarantees, idempotent consumers, ordering, dead-letter handling, and event sourcing.

fastapi

Use when building APIs with FastAPI. Covers dependency injection, Pydantic v2 validation, async database access, authentication, background tasks, and testing.

graphql

Use when designing or operating a GraphQL API. Covers schema design, resolver performance and DataLoader batching, query cost limiting, error handling, and federation.

microservices

Use when decomposing a system into services, or deciding whether to. Covers service boundaries, inter-service communication, distributed data, saga patterns, and the operational cost of the network.

nestjs

Use when building NestJS services. Covers module structure, providers and scopes, validation pipes, guards and interceptors, TypeORM/Prisma integration, and testing.

realtime-websockets

Use when building realtime features with WebSockets or Server-Sent Events. Covers protocol choice, connection lifecycle, reconnection and backfill, scaling across instances, and backpressure.

Testing & QA Listed

spring-boot

Use when building Spring Boot services. Covers dependency injection, transaction boundaries, JPA performance, configuration, validation, and testing slices.

Code & Development Listed

competitive-analysis

Use when analyzing competitors. Covers gathering evidence rather than opinion, positioning and pricing analysis, identifying real differentiation, and producing a conclusion that changes a decision.

lead-research

Use when researching prospects or accounts before outreach. Covers qualification, finding the trigger event, identifying the right contact, and producing a brief that makes outreach specific rather than generic.

product-analysis

Use when analyzing a product's performance or deciding what to build. Covers metric selection, funnel and retention analysis, distinguishing signal from noise, and prioritizing on evidence.

resume

Use when writing or tailoring a CV or resume. Covers evidence over adjectives, tailoring to a specific role without lying, structure, and the specific patterns that get a resume discarded.

Testing & QA Listed

user-research

Use when planning, running, or synthesizing user research. Covers interview questions that produce evidence rather than opinions, usability testing, and turning transcripts into findings.

data-modeling

Use when designing a schema. Covers normalization and when to break it, choosing keys, modeling time and history, soft deletes, multi-tenancy, and schema decisions that are expensive to reverse.

data-quality

Use when validating a dataset or building quality checks into a pipeline. Covers profiling, schema and constraint validation, freshness and completeness checks, anomaly detection, and failing a pipeline correctly.

database-performance

Use when a database is the bottleneck. Covers finding the expensive queries, index strategy, lock contention, connection saturation, and the schema decisions that make queries fast or impossible.

pandas

Use when analyzing or transforming tabular data in Python. Covers vectorized operations, memory-efficient dtypes, correct joins, groupby patterns, and avoiding the silent errors pandas makes easy.

postgres

Use when working with PostgreSQL specifically. Covers indexing, MVCC and vacuum, connection pooling, partitioning, JSONB, replication, and the operational realities that separate Postgres from generic SQL.

spark

Use when building distributed data pipelines with Apache Spark. Covers partitioning, shuffles, skew, joins, caching, and reading the Spark UI to find why a job is slow.

brand-guidelines

Use when applying or documenting a brand identity. Covers extracting a system from existing assets, documenting colour, type, and voice, and applying a brand consistently across new work.

generative-art

Use when creating algorithmic or generative visual art. Covers composition through code, controlled randomness, colour systems, and building work that is varied without being arbitrary.

theming

Use when building a theme or theming system for an interface. Covers token architecture, dark mode, contrast preservation across themes, and generating a coherent palette from a seed colour.

ui-review

Use when critiquing an interface design. Covers hierarchy, affordance, consistency, states, and delivering feedback that changes the design rather than describing taste.

visual-design

Use when creating posters, covers, social graphics, or other static visual designs. Covers composition, typographic hierarchy, colour, and producing original work rather than pastiche.

DevOps & Infrastructure Listed

architecture-decisions

Use when a significant technical decision needs to be recorded. Produces Architecture Decision Records that capture context, options, the decision, and its consequences — so the reasoning survives the people who made it.

bug-fix-protocol

Use when fixing a reported bug end to end. Enforces a disciplined sequence — reproduce, write a failing test, fix minimally, verify, prevent recurrence — and blocks the common failure mode of patching symptoms.

cli-development

Use when building command-line tools. Covers argument design, exit codes, streams and piping, progress output, configuration precedence, and behavior that respects the shell.