robertoatila
- 311 skills
- 0 followers
- 9 repo stars
- 3 days ago last updated
- ▌ Vllm Paged Attention High Throughput Serving · robertoatilaServes LLMs at scale using continuous batching and PagedAttention memory management.
- ▌ Brandkit Design System 2 · robertoatilaPremium brand-kit image generation skill for creating high-end brand-guidelines boards, logo systems, identity decks, and visual-world presentations. Trained for minimalist, cinematic, editorial, dark-tech, luxury, cultural, security, gaming, developer-tool, and consumer-app brand systems. Optimized for intentional logo concepting, refined composition, sparse typography, strong symbolic meaning, premium mockups, art-directed imagery, and flexible grid layouts.
- ▌ Github Issue Pr Triage 2 · robertoatilaTriage incoming GitHub issues and Pull Requests using GitHub CLI (gh) in read-only analysis mode. Prioritizes bugs, identifies regressions, maps duplicates, and drafts structured resolution proposals. Triggers: github triage, issue triage, pr triage, triage issues, triage github, bug triage.
- ▌ Hyperplan Orchestrator 2 · robertoatilaStructure complex, multi-phase technical projects into deterministic, file-backed implementation plans with explicit verification criteria and rollback safeguards. Triggers: hyperplan, plan project, multi-phase plan, plan architecture, implementation plan, orchestrate plan.
- ▌ JSON Canvas Visualizer 2 · robertoatilaCreate and edit JSON Canvas files (.canvas) with nodes, edges, groups, and connections. Use when working with .canvas files, creating visual canvases, mind maps, flowcharts, or when the user mentions Canvas files in Obsidian.
- ▌ Systematic Refactoring 2 · robertoatilaIntelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.
- ▌ UI Redesign Modernizer 2 · robertoatilaUpgrades existing websites and apps to premium quality. Audits current design, identifies generic AI patterns, and applies high-end design standards without breaking functionality. Works with any CSS framework or vanilla CSS.
- ▌ Agentic Idea Refinement 2 · robertoatilaRefines raw ideas into sharp, actionable concepts through structured divergent and convergent thinking. Use when an idea is still vague, when you need to stress-test assumptions before committing to a plan, or when you want to expand options before converging on one. Triggers on "ideate", "refine this idea", or "stress-test my plan".
- ▌ Agentic Loop Controller 2 · robertoatilaGoal-like loop that uses ultrawork mode to decompose work into systematic, evidence-bound steps.
- ▌ Deep Project Scaffolder 2 · robertoatila(builtin) Initialize hierarchical AGENTS.md knowledge base Triggers: init-deep, deep-project-scaffolder, verify, tool.
- ▌ Deep Technical Research 2 · robertoatilaMaximum-saturation research orchestration: ALWAYS proposes the final materials first (PDF+DOCX default), then parallel explore+librarian swarms across codebase, web, official docs, and OSS repos — max-roster teammode when the harness has it — with live journaling, a recursive EXPAND loop driven by leads workers return in message text, empirical verification by running code, and a cited synthesis with charts/Mermaid/assets behind a mandatory visual-QA gate. ACTIVATES ONLY on an explicit user demand for research — the word 'ulw-research' in any prefix form, any 'ulw' research wording including combined invocations like 'mass ulw research', 'ultradebate' or 'hyperdebate' research requests, or an explicit request for research / deep research / an ultra-precise investigation, in any language. Never self-activates for ordinary questions, debugging, or implementation context-gathering. While active it overrides exploration-bounding defaults: exhaustive coverage is the goal.
- ▌ Obsidian CLI Controller 2 · robertoatilaInteract with Obsidian vaults using the Obsidian CLI to read, create, search, and manage notes, tasks, properties, and more. Also supports plugin and theme development with commands to reload plugins, run JavaScript, capture errors, take screenshots, and inspect the DOM. Use when the user asks to interact with their Obsidian vault, manage notes, search vault content, perform vault operations from the command line, or develop and debug Obsidian plugins and themes.
- ▌ Obsidian Database Bases 2 · robertoatilaCreate and edit Obsidian Bases (.base files) with views, filters, formulas, and summaries. Use when working with .base files, creating database-like views of notes, or when the user mentions Bases, table views, card views, filters, or formulas in Obsidian.
- ▌ PDF Document Processing 2 · robertoatilaComprehensive PDF manipulation toolkit for extracting text and tables, creating new PDFs, merging/splitting documents, and handling forms. When Claude needs to fill in a PDF form or programmatically process, generate, or analyze PDF documents at scale.
- ▌ Security Research Audit 2 · robertoatilaAudit and research security vulnerabilities, CVEs, and dependency risks across project components in strict read-only mode. Identifies exposed secrets, insecure configurations, and vulnerable third-party dependencies with actionable remediation evidence. Triggers: security research, vulnerability audit, cve audit, security audit, dependency scan, audit security.
- ▌ Agent Browser Automation 2 · robertoatilaAutomates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.
- ▌ Agent Environment Doctor 2 · robertoatilaDiagnose LazyCodex and Codex CLI installation health against the latest sources. Use whenever the user asks for a doctor or health check, says LazyCodex, lazycodex-ai, omo-codex, or Codex behaves oddly after an install, update, or config change, suspects a stale, drifted, or broken setup, or wants the local install audited and compared with the latest LazyCodex and Codex code. Triggers: lcx-doctor, agent-environment-doctor, verify, tool.
- ▌ Qdrant Vector Database Hnsw Indexing · robertoatilaHigh-performance vector similarity search engine with HNSW indexing and metadata filtering.
- ▌ Ast Grep Search 2 · robertoatilaUse ast-grep (sg) for AST-aware code search and rewrite across 25 languages. Trigger for structural code matching or deterministic codemods: find every function/call/class/import shaped like X, rewrite console.log to logger.info, strip `as any`, migrate require() to import, find empty catch blocks or missing await, and scan/apply YAML rules. Prefer this over rg/grep when the target is syntax shape rather than text; use rg for string contents, comments, filenames, or regex-style byte searches.
- ▌ Codex Plugin QA 2 · robertoatilaQA and verify OpenAI Codex plugins and agent hooks in strict isolation (isolated CODEX_HOME + local mock model). Prevents real ~/.codex pollution, asserts hook/started and hook/completed events, and validates TUI/app-server flows with cross-platform fallbacks. Triggers: codex qa, qa codex, codex-plugin-qa, test codex plugin, verify codex hook, codex app-server, isolated CODEX_HOME.
- ▌ Tech Debt Audit 3 · robertoatilaAudit, quantify, and categorize architectural, technical, and testing debt across a codebase. Produces a structured Tech Debt Scorecard with prioritized remediation backlog. Triggers: tech debt, tech debt audit, code debt, audit debt, architectural debt, refactoring backlog.
- ▌ Subagent Task QA 2 · robertoatilaQA and verify sub-agent task engines, DAG orchestration, and multi-agent coordination in strict isolation with deterministic evidence tracking. Rejects path traversal and stray evidence roots, preserves user sandbox integrity, and validates terminal task states. Triggers: subagent qa, qa subagent, subagent-task-qa, task dag qa, verify agent task, multi-agent e2e, agent evidence path.
- ▌ Speculative Decoding Draft Verification · robertoatilaAccelerates LLM inference via speculative decoding, draft verification, and multi-token prediction.
- ▌ Github Bug Reporter 2 · robertoatilaCreate a high-signal bug issue or PR in the repo that owns the defect. Use this whenever the user asks to report, file, open, or triage a LazyCodex, lazycodex-ai, omo-codex, Codex plugin, or upstream Codex CLI bug, especially when they need source-backed root cause, reproduction steps, fix guidance, and GitHub routing.
- ▌ Opencode Runtime QA 2 · robertoatilaQA and verify OpenCode runtime configurations, plugin hooks, tool definitions, and session state in strict isolation. Prevents workspace pollution and verifies deterministic agent turns. Triggers: opencode qa, qa opencode, opencode-runtime-qa, test opencode plugin, verify opencode.
- ▌ Soak Stress Testing 2 · robertoatilaGenerate a soak test protocol for extended play sessions. Defines what to observe, measure, and log during long play sessions to surface slow leaks, fatigue effects, and edge cases that only appear after sustained play. Primarily used in Polish and Release phases. Triggers: soak-test, soak-stress-testing, execute, inspect.
- ▌ Flash Attention Kernel Tiling Acceleration · robertoatilaOptimizes transformer attention kernels for speedup and memory reduction using FlashAttention.
- ▌ Data Science Toolkit 2 · robertoatilaExpert data processing with a hybrid engine strategy: resident-kernel engines first - DuckDB plus a resident Python stack (Polars/numpy/matplotlib) in persistent js/py eval kernels where the harness has them, bun/uv one-shots elsewhere - and per-action placement judgment (in-memory vs streaming vs remote-in-place). Triggers: 'analyze the data', 'what is in this CSV/parquet/json', 'summarize this', 'group by', 'filter rows', 'sort by', 'join these files', 'merge datasets', 'time series trend', 'compare yesterday and today', 'distribution/histogram', 'correlation', 'clean duplicates', 'handle missing values', 'dataset larger than RAM', 'SQL query on files', 'DataFrame operations', 'chart/plot this data', DuckDB vs Polars selection, quick data exploration CLI. NOT for plain text/code inspection, configs, or tiny inline math.
- ▌ Deadcode Elimination 2 · robertoatilaSafely identify and prune unused functions, orphan modules, dead exports, and obsolete dependencies. Enforces compilation and test verification before each deletion batch. Triggers: remove deadcode, deadcode elimination, prune unused code, delete dead code, cleanup exports.
- ▌ Git Advanced Mastery 2 · robertoatilaMUST USE whenever a task needs a commit or git-history investigation. Covers atomic commits, staging, commit-message style, rebase, squash, fixup/autosquash, blame, bisect, reflog, git log -S/-G, and questions like who wrote this or when was this added. Do not use for ordinary code edits unless the user asks for git work. Triggers: git-master, git-advanced-mastery, audit, scan.
- ▌ Lsp Diagnostic Setup 2 · robertoatilaConfigure a Language Server (LSP) for a specific language so editor/agent tooling — diagnostics, go-to-definition, find-references, rename — works. Use when you need to: configure LSP, lsp setup, set up or install a language server, fix 'no LSP server configured' / 'server not installed', choose between servers (basedpyright vs pyright vs ty vs ruff), or wire .codex/lsp-client.json / .opencode/lsp.json. 언어서버 설정. Routes by file extension to references/<language>/README.md for the exact builtin server, per-OS install commands (macOS/Linux/Windows), config snippets for both config files, initialization options, alternatives, and troubleshooting. Ships scripts: detect-lsp.ts (scan a project for languages + each server's install/config status) and verify-lsp.ts (run a real diagnostics roundtrip). Covers typescript, python, go, rust, c/c++, java, kotlin, c#/razor, swift, ruby, php, dart, elixir, zig, lua, bash, yaml, terraform, haskell, julia. Triggers: lsp-setup, lsp-diagnostic-setup, audit, scan.
- ▌ Multi Agent Teammode 2 · robertoatilaCodex-only team orchestration: run a named team of cooperating Codex workers with durable, script-managed state. MUST USE when the user asks Codex to create, run, coordinate, inspect, archive, or delete a team of agents/threads/sessions, or to work on something as a team in parallel. FIRST inspects the active tool surface (checking tool_search for deferred tools) and tells the user the route: native MultiAgentV2 agents (flat spawn_agent with task_name) when available, Codex App threads as the fallback, or a plain-subagent split when neither set exists. The main session is always the leader; members are defined by a concrete part, ownership area, or perspective - never a vague job role; a bundled cross-platform script writes the .omo/teams state plus an auto-generated member field manual. Use a team when the work is not perfectly isolated but parallelizing helps; use plain subagents when scope is perfectly isolated or the goal is ambiguous. Triggers: team mode, teammode, make a team, run as a team, team of age
- ▌ Pr Review Resolution 2 · robertoatilaSystematically analyze, address, and resolve review comments on pull requests. Ensures full test validation and structured resolution replies without destructive git actions. Triggers: work with pr, pr review, resolve pr comments, address review, fix pr comments, pr feedback.
- ▌ Visual Regression QA 2 · robertoatilaMUST USE after building/changing any UI or when asked whether a page, component, or TUI looks right. Rigorous visual QA across web/page, terminal, and paginated-document surfaces. Prefer browser:control-in-app-browser for unauthenticated browser/page QA in Codex, then Playwright/agent-browser/dev-browser. Captures screenshot/TUI evidence with bundled diff scripts, runs design-system/functional and visual-fidelity/CJK reviewer passes, then synthesizes a good/bad verdict. Triggers: visual QA, screenshot/pixel diff, UI looks wrong, reference fidelity, design system check, responsive check, CJK text clipping, TUI alignment, box-drawing drift, PDF page check, deck review, blank or near-empty page, wrong page break.
- ▌ Coding Agent Sessions 2 · robertoatilaMUST USE when asked to find, read, list, search, inspect, fetch, export, or reconstruct coding-agent sessions across Codex, Claude Code/Desktop, OpenCode, OMO/Senpi/pi, oh-my-pi (omp), gajae-code (gjc), OpenClaw, Factory Droid, Amp, Gemini/Kimi/Qwen CLIs, Codebuff, Roo/Kilo/Cline, Kodu, Cursor CLI, Aider, Aside browser-agent sessions, or unknown local agent logs. Covers transcripts, session IDs, rollout JSONL, state SQLite, Claude projects/pre-compact histories, OpenCode messages/parts, child/subagent linkage, cwd/model/time/token filters, archives, and cost clues. Expands fuzzy recall into parallel query lanes and first probes known stores so absent platforms are skipped cheaply. Triggers: coding agent sessions, Codex/Claude/OpenCode/OMO/Senpi/pi/oh-my-pi/omp/gajae-code/gjc/OpenClaw/Droid/Amp/Kodu/Cursor/Aider/Aside sessions, transcript search, session history, session ID, read transcript, token usage, subagent sessions, what did I do yesterday, did we already do this.
- ▌ Rea · robertoatilaReverse engineer anything with agents, from app behavior down to native binaries.
- ▌
- ▌ Fastapi · robertoatilaFastAPI framework, high performance, easy to learn, fast to code, ready for production
- ▌
- ▌
- ▌ Freecodecamp · robertoatilafreeCodeCamp.org's open-source codebase and curriculum. Learn math, programming, and computer science for free.
- ▌ Developer Roadmap · robertoatilaInteractive roadmaps, guides and other educational content to help developers grow in their careers.
- ▌ System Design Primer · robertoatilaLearn how to design large-scale systems. Prep for the system design interview. Includes Anki...
- ▌ Jarvis Pc Operator Automation · robertoatilaAutomates Windows PC operations, PowerShell diagnostics, Free AI routing, and Mark-LIV HUD.
- ▌ Dspy Programmatic Prompt Compilation · robertoatilaCompiles declarative LM programs, modular RAG systems, and self-optimizing pipelines with DSPy.
- ▌ Sglang Radix Attention Runtime · robertoatilaHigh-throughput LLM serving with RadixAttention prefix caching and structured decoding.
- ▌ Software Construction Patterns · robertoatilaApplies resilient software design patterns, clean architecture, and modular domain modeling.
- ▌ Unsloth Fast Kernel Finetuning · robertoatilaFast, memory-efficient LLM fine-tuning using Unsloth LoRA/QLoRA optimized kernels.
- ▌ Bigcode Harness Code Evaluation · robertoatilaEvaluates code generation models across HumanEval, MBPP, and multi-language benchmarks.
- ▌
- ▌ Cosmos Physical AI World Policy · robertoatilaEvaluates NVIDIA Cosmos physical AI policies in simulated robotics environments.
- ▌ Deckgl Geospatial Visualization · robertoatilaHigh-performance WebGL2 and WebGPU large-scale reactive data layer and geospatial visualization engine.
- ▌ Llava Visual Instruction Tuning · robertoatilaMultimodal visual instruction tuning combining CLIP vision encoders with large language models.
- ▌ Mlflow Model Registry Lifecycle · robertoatilaTracks ML experiments, versioning, artifacts, and production deployments with MLflow.
- ▌ Qdrant Vector Storage Retrieval · robertoatilaVector similarity search and retrieval-augmented generation engine with hybrid filtering.
- ▌ Segment Anything Visual Masking · robertoatilaZero-shot visual segmentation and mask generation using Segment Anything (SAM).
- ▌ Structured Pydantic LLM Outputs · robertoatilaExtracts structured data from LLMs with Pydantic type validation and retries.
- ▌ Vision Language Action Robotics · robertoatilaFine-tunes and serves OpenVLA robot manipulation policies from vision-language inputs.
- ▌ Cellxgene Census Tissue Querying · robertoatilaQueries CZ CELLxGENE Census single-cell and spatial transcriptomics datasets programmatically.
- ▌ Gguf Tensor Format Serialization · robertoatilaQuantizes and serves models in GGUF format for efficient CPU/GPU inference.
- ▌ Nemo Evaluator Safety Guardrails · robertoatilaEvaluates LLM safety guardrails, jailbreak resistance, and toxicity using NVIDIA NeMo.
- ▌ Gene Regulatory Network Inference · robertoatilaInfers gene regulatory networks from bulk and single-cell transcriptomics data.
- ▌ Neuroimaging Bids Standardization · robertoatilaOrganizes neuroscience and biomedical datasets according to Brain Imaging Data Structure standards.
- ▌ Polars Streaming Dataframe Engine · robertoatilaHigh-performance streaming DataFrame engine with lazy query optimization and Apache Arrow integration.
- ▌ Pydicom Medical Imaging Radiology · robertoatilaReads, inspects, validates, and transforms medical DICOM datasets and pixel data.
- ▌ Qiskit Quantum Circuit Algorithms · robertoatilaDesigns, simulates, transpiles, and executes quantum circuits using IBM Qiskit.
- ▌ Skypilot Multicloud Orchestration · robertoatilaMulti-cloud compute orchestration and automatic cost optimization for distributed ML workloads.
- ▌ Speculative Decoding Acceleration · robertoatilaAccelerates LLM inference throughput using speculative decoding and Medusa multi-head draft prediction.
- ▌ Tmux Session Terminal Multiplexer · robertoatilaControls tmux terminal sessions programmatically for interactive CLI automation.
- ▌ Analytical Method Validation · robertoatilaValidates analytical procedures, assay precision, specificity, linearity, and regulatory compliance standards.
- ▌ LLM Model Merging Techniques · robertoatilaMerges multiple fine-tuned LLMs using mergekit without retraining or compute overhead.
- ▌ Single Cell Anndata Genomics · robertoatilaManipulates single-cell genomics matrices and metadata using AnnData structures.
- ▌ Vllm High Throughput Serving · robertoatilaHigh-throughput LLM serving with PagedAttention, vLLM continuous batching, and quantization.
- ▌ API Contract Interface Design · robertoatilaDesigns stable API contracts, boundary types, and robust public interfaces.
- ▌ Architecture Decision Records · robertoatilaDocuments significant technical decisions, architectural context, alternatives, and consequences via ADRs.
- ▌ Constraint Driven Development · robertoatilaEstablishes written quality contracts in CONSTRAINTS.md to enforce standards and stop.
- ▌ Developer Onboarding Workflow · robertoatilaAutomates developer environment onboarding, prerequisites checks, and project tour.
- ▌ Git Unpublished Changes Audit · robertoatilaAudits unpublished commits and tags against release registries across monorepos.
- ▌ Neural Model Pruning Sparsity · robertoatilaPrunes neural network weights for sparsity and inference acceleration using Wanda/SparseGPT.
- ▌ Presentation Deck Engineering · robertoatilaDesigns, edits, and structures presentation decks and slides with visual hierarchy.
- ▌
- ▌ Developer Productivity Metrics · robertoatilaAnalyzes developer coding patterns, chat logs, productivity bottlenecks, and learning resources.
- ▌ Long Context Rope Scaling Eval · robertoatilaExtends transformer context windows using RoPE, YaRN, ALiBi, and position interpolation.
- ▌ Responsive Layout Architecture · robertoatilaArchitects responsive app shell layouts, collapsible sidebars, and adaptive containers.
- ▌ Ast Grep Search · robertoatilaAST-aware code search and structural rewrites across 25 programming languages.
- ▌ Codex Plugin QA · robertoatilaQAs and tests Codex plugins, lifecycle hooks, and tool handlers in isolation.
- ▌ Tech Debt Audit · robertoatilaAudits architectural and test debt, generating prioritized remediation scorecards.
- ▌ Subagent Task QA · robertoatilaQAs subagent DAG orchestration, task isolation, and deterministic evidence trails.
- ▌ Openrouter AI Sdk · robertoatilaMulti-model AI streaming and fallback router across 200+ LLMs via OpenRouter.
- ▌ Free AI Apis Router · robertoatilaResilient fallback router and dispatcher for free-tier and sovereign local AI.
- ▌ Github Bug Reporter · robertoatilaFormats and files high-signal bug reports with minimal reproducible testcases.
- ▌ Opencode Runtime QA · robertoatilaQAs and verifies OpenCode runtime configs, plugin hooks, and isolated session state.
- ▌ Soak Stress Testing · robertoatilaExecutes extended soak and stress tests to surface memory leaks and fatigue.
- ▌
- ▌ Browser Devtools Inspector · robertoatilaInspects DOM, CSS styles, console logs, and network traffic via Chrome DevTools.
- ▌ LLM Eval Benchmark Harness · robertoatilaRuns promptfoo evaluation suites, model-graded rubrics, and regression test matrices.
- ▌ Mixture Of Experts Routing · robertoatilaTrains and serves Mixture of Experts (MoE) models with DeepSpeed and HuggingFace.
- ▌ Ultrawork Execution Engine · robertoatilaGoal-oriented execution loop decomposing complex projects into verifiable, systematic milestones.
- ▌
- ▌ Awq Activation Quantization · robertoatilaActivation-aware weight quantization for 4-bit LLM compression with low latency.