Loading...
Loading...
Found 172 Skills
Academic literature orientation skill that searches papers via Consensus, builds a strategic search plan using PICO (default) or SPIDER / Decomposition / hybrid as fallbacks, and synthesizes findings into a professionally formatted Word document (.docx) research guide. Grill-me intake (research question specificity + framework hint + tentative depth) before the recon search; a second forcing checkpoint after Phase 2 confirms framework + sub-areas + depth before searches consume budget. Configurable depth (5/10/20 queries) controls coverage vs. speed. Output is a 'launching pad' — not a finished review, but an orientation guide that lets a researcher dive in confidently. Triggers: 'litreview on [topic]', 'literature review on [topic]', 'I'm starting a literature review on X', 'I'm writing a paper on X', 'help me research X', 'I'm doing research on X', 'can you help me research X'. Do NOT trigger for single one-off paper searches where the user just wants a quick list — that's a plain Consensus search.
Gerente do ciclo de vida de stories e orquestrador de handoffs no SynkOS. Use esta skill quando o usuário pedir para decompor um épico em stories, criar stories com critérios de aceite, fazer backlog grooming, planejar sprint, orquestrar handoffs entre roles (architect → dev → qa), ou fazer perguntas como "quebre esse épico em stories", "crie a story para X", "o backlog está priorizado?", "faça o checkpoint da story Y", "orquestre o handoff para QA". Ative também para resolver dependências entre stories, escalar stories bloqueadas, e garantir que cada story tem ownerRole e reviewRole definidos antes de entrar em implementação.
Autonomous research agent that reads RESEARCH.md, infers what's needed, dynamically adjusts TODOs, and delegates to the right skill. Supports opt-in BFS mode for autonomous design space search. Respects a configurable supervision policy (presets: manual / checkpointed / autonomous / wild) governing notifications, approval gates, resource limits, and idea-change handling. Proactively surfaces gaps and asks before acting. Trigger phrases: "start research", "continue project", "what's next?", "explore design space", "autoresearch".
This skill should be used at natural checkpoints (after completing complex tasks, at session end, or when friction occurs) to reflect on skill and process execution and identify targeted improvements. Use when experiencing confusion, repeated failures, or discovering new patterns that should be codified into skills for smoother future operation.
Use this skill when the user wants to manage remote Sprites from their local machine — listing sprites, executing commands, managing checkpoints, transferring files, controlling network policy, or coordinating work across multiple sprites.
Track long-horizon objectives across multiple sessions with milestone checkpoints, progress persistence, and drift detection
Builds, deploys, manages, debugs, configures, and optimizes serverless applications on AWS using Lambda, API Gateway, Step Functions, EventBridge, and SAM/CDK. Covers cold starts, CORS debugging, event source mappings, troubleshooting, concurrency, SnapStart, Powertools, function URLs, EventBridge Scheduler, Lambda layers, Durable Functions, durable execution, checkpoint-and-replay, and production readiness. Use when the user mentions Lambda, API Gateway, Step Functions, SAM templates, CDK serverless stacks, DynamoDB stream triggers, SQS event sources, cold starts, timeouts, 502/504 errors, throttling, concurrency, CORS, Powertools, Durable Functions, durable execution, checkpoint-and-replay, or any event-driven architecture on AWS, even if they don't say "serverless." Do NOT use for EC2, ECS/Fargate containers, or Amplify hosting.
Serve a quantized or unquantized LLM checkpoint as an OpenAI-compatible API endpoint using vLLM, SGLang, or TRT-LLM. Use when user says "deploy model", "serve model", "start vLLM server", "launch SGLang", "TRT-LLM deploy", "AutoDeploy", "benchmark throughput", "serve checkpoint", or needs an inference endpoint from a HuggingFace or ModelOpt-quantized checkpoint. Do NOT use for quantizing models (use ptq) or evaluating accuracy (use evaluation).
Code instrumentation for timing workloads. Two scenarios: (1) Training loop — inject manual timing to report per-iteration latency, throughput (samples/sec), and data load time. (2) Standalone kernel/op — write CUDA event timing code with warmup, per-iteration statistics, and anti-pattern avoidance. Also covers NVTX annotation for labeling profiler timelines. NOT for: running or analyzing profiler tools (nsys, ncu, Nsight Systems, Nsight Compute), writing kernels (Triton, CuTe, CUDA), applying optimizations (CUDA Graphs, gradient checkpointing, fusion), or interpreting roofline/SOL% metrics. Triggers: "measure throughput", "benchmark this function", "time my training loop", "samples per second", "NVTX annotate", "instrument my dataloader", "data load time", "kernel timing", "how do I time".
Anti-footgun protocol for AI-assisted coding. Always active during coding tasks to enforce simplicity-first thinking, surface assumptions, and prevent scope creep. Explicit checkpoints available via "cg pre", "cg post", "cg simplify". Triggers on: any coding task, code review requests, refactoring, or when user says "cg" or "check".
LangGraph state-machine design and debugging for `StateGraph`, node/edge routing, checkpoints, `interrupt`, and HITL flows. Use when building or troubleshooting graph-based agents with conditional edges and thread state.
Save progress and generate a continuation prompt. Updates PRD status markers, captures git state, and writes checkpoint.md for the next session.