aleonsa
- 24 skills
- 0 followers
- 14 hours ago last updated
- ▌ Graphify · aleonsa bundleUse for any question about a codebase, its architecture, file relationships, or project content — especially when graphify-out/ exists, where the question should be treated as a graphify query first. Turns any input (code, docs, papers, images, videos) into a persistent knowledge graph with god nodes, community detection, and query/path/explain tools.
- ▌ Caveman Help · aleonsa bundleQuick-reference card for all caveman modes, skills, and commands. One-shot display, not a persistent mode. Trigger: /caveman-help, "caveman help", "what caveman commands", "how do I use caveman".
- ▌ Caveman Compress · aleonsa bundleCompress natural language memory files (CLAUDE.md, todos, preferences) into caveman format to save input tokens. Preserves all technical substance, code, URLs, and structure. Compressed version overwrites the original file. Human-readable backup saved as FILE.original.md. Trigger: /caveman-compress FILEPATH or "compress memory file"
- ▌ Iterative Retrieval · aleonsaPattern for progressively refining context retrieval to solve the subagent context problem
- ▌ GCP Debug · aleonsaDiagnose latency, errors, freezes and scaling issues across Cloud Run + Apigee + Pub/Sub on GCP. Use when the user reports production incidents, latency spikes, partial outages, or asks to "see logs / metrics" in a GCP project.
- ▌ Fastapi Tdd · aleonsaFastAPI testing with TDD — pytest-asyncio, httpx AsyncClient, async SQLAlchemy fixtures, polyfactory, mocking external services, and coverage targets.
- ▌ Latex Beamer · aleonsaLaTeX compilation and Beamer presentation patterns for macOS with TeX Live. Covers pdflatex, latexmk, bibtex, error diagnosis, and Beamer slide authoring.
- ▌ MCP Patterns · aleonsaMCP (Model Context Protocol) server patterns — FastMCP server design, tool/resource/prompt definitions, authentication, testing, and deployment.
- ▌ RAG Patterns · aleonsaRAG architecture patterns — document ingestion, chunking strategies, embedding models, vector stores, hybrid retrieval, re-ranking, advanced techniques (HyDE, multi-query), and evaluation with RAGAS.
- ▌ Firebase Auth · aleonsaFirebase Authentication patterns — token verification in FastAPI, custom claims for RBAC, multi-tenant auth, ID token validation, service account setup, and frontend Firebase SDK integration.
- ▌ GCP Serverless · aleonsaGCP serverless deployment patterns — Cloud Run, Cloud Tasks, Secret Manager, Artifact Registry, Cloud Storage, IAM, and cost-aware architecture for production Python services.
- ▌ Twilio Extract · aleonsa bundleQuery Twilio Conversations, Studio Flows, and Monitor/Debugger via the REST API — find conversations by phone/WhatsApp number, read messages and attributes, inspect Studio Flow executions, list debugger alerts. Use when the user wants to look up/inspect/debug Twilio conversations, or check a phone number's history in Twilio. Not for instrumenting a channel adapter — that's a coding task.
- ▌ LLM Engineering · aleonsaLLM engineering patterns — prompt design, structured outputs with instructor, tool use, context management, model selection/routing, cost optimization, streaming, retry logic, and observability.
- ▌ Autonomous Loops · aleonsaPatterns and architectures for autonomous Claude Code loops — from simple sequential pipelines to RFC-driven multi-agent DAG systems.
- ▌ Fastapi Patterns · aleonsaFastAPI architecture patterns, async SQLAlchemy ORM, Pydantic schemas, dependency injection, service layer, background tasks, and production-grade API design.
- ▌ Fastapi Security · aleonsaFastAPI security best practices — JWT authentication with python-jose, OAuth2, role-based access, CORS, rate limiting, security headers, and input validation.
- ▌ Langfuse Extract · aleonsa bundleExtract sessions, traces, observations, prompts, and scores from Langfuse via its public API. Use when the user wants to pull/export/analyze Langfuse data (not for instrumenting an app to send traces — see langfuse-observability for that).
- ▌ Intercom Patterns · aleonsaIntercom integration patterns — webhook event handling, Conversations API, Contacts API, custom inbox channels, operator bot flows, and multi-tenant workspace management.
- ▌ Multi Tenant Saas · aleonsaMulti-tenant SaaS patterns — tenant isolation strategies, row-level security, tenant-aware SQLAlchemy models, RBAC, subdomain routing, and Alembic migrations for shared schema.
- ▌ Kubernetes Patterns · aleonsaKubernetes production patterns — Deployments, Services, Ingress, ConfigMaps, Secrets, RBAC, HPA, health probes, resource limits, and Helm chart structure for containerized Python/Node services.
- ▌ Fastapi Verification · aleonsaVerification loop for FastAPI projects: Alembic migration checks, linting, async tests with coverage, security scans, and deployment readiness before release or PR.
- ▌ Continuous Learning V2 · aleonsa bundleInstinct-based learning system that observes sessions via hooks, creates atomic instincts with confidence scoring, and evolves them into skills/commands/agents. v2.1 adds project-scoped instincts to prevent cross-project contamination.
- ▌ Langfuse Observability · aleonsaLLM observability with Langfuse — tracing LLM calls, spans, prompt versioning, cost tracking, evaluation datasets, and multi-tenant trace isolation in FastAPI/Python applications.
- ▌ Cuda Kernel Optimization · aleonsaCUDA kernel development and GPU optimization patterns — memory hierarchy, occupancy tuning, coalescing, shared memory tiling, warp-level ops, and profiling with Nsight Compute. Use when writing or optimizing CUDA C++ kernels.