Agent Skills by dhumitech
Showing 46 vetted skills indexed across 1 GitHub repositories.
fastapi-pro
Create a production-grade FastAPI foundation with async SQLAlchemy and OpenAPI documentation.
ai-integration
Expose trained ML models as production APIs with OpenAPI specs and Docker containers.
python-pro
Initialize Python 3.12+ projects with typing, formatting, and packaging tooling.
async-python-patterns
Automate asynchronous Python programming with asyncio patterns for non-blocking applications.
python-patterns
Identify and apply idiomatic Python patterns for code quality and maintainability.
observability-engineer
Design observability signals, dashboards, and alerting for production services.
deployment-procedures
Plan and execute safe deployment rollouts with blue-green, canary, and rollback procedures.
docker-expert
Audit Dockerfiles and implement multi-stage builds for optimized images.
database-design
Design relational database schemas and select ORMs with migration planning.
embedding-strategies
Compare and select embedding models for retrieval pipelines.
postgres-best-practices
Provide PostgreSQL best-practice guidelines for query performance and schema design.
data-engineer
Automate design and deployment of scalable ETL/ELT and streaming data pipelines.
vector-database-engineer
Optimize vector database architectures for semantic search in RAG applications.
data-pipeline
Automate data preprocessing, cleaning, and transformation for ML-ready datasets.
auth-implementation-patterns
Implement JWT, OAuth2, and RBAC/ABAC authentication patterns for web applications.
backend-security-coder
Apply defensive programming to prevent injection, SSRF, and deserialization attacks in backend code.
security-auditor
Identify and remediate security vulnerabilities across codebases, architectures, and deployments.
api-security-best-practices
Integrate authentication, authorization, input validation, and rate limiting for REST, GraphQL, and WebSocket APIs.
documentation
Automate creation and maintenance of AI system documentation with templates and validation prompts.
doc-coauthoring
Guide structured collaborative drafting of technical documents through three workflow stages.
langfuse
Instrument LLM pipelines for latency, token usage, and hallucination tracking with Langfuse.
langgraph
Build graph-based AI agent workflows with conditional routing and persistent state.
rag-implementation
Implement end-to-end RAG pipelines for ingestion, embedding, retrieval, and generation.
prompt-engineering
Design robust prompts for LLMs with system prompts, few-shot patterns, and chain-of-thought reasoning.