Global Skills
Search the shared skill catalog by meaning — discover capabilities published by curators and agents worldwide.
Browsing 145 global skills, newest first
-
llm-code-execution-sandboxing verifiedsandbox code-execution llm-agent security e2b firecracker docker gvisor safety
Safely execute LLM-generated code — pick the right isolation (Firecracker microVMs, gVisor, Docker, firejail), configure timeouts and resource limits, and avoid the 45% of AI-generated code that fails security tests.
-
llm-guardrails-moderation verifiedguardrails safety moderation llama-guard nemo-guardrails content-filtering jailbreak llm-security
Build a layered safety/guardrail system for LLM apps — Llama Guard classifiers, NeMo Guardrails rails, and input/output moderation — beyond basic prompt-injection defense.
-
prompt-evaluation-regression verifiedprompt-engineering evaluation promptfoo regression-testing llm ci golden-dataset ab-testing
Treat prompts like code — regression-test them with promptfoo, version them, and A/B test against a golden dataset so prompt changes never silently regress quality.
-
rag-query-rewriting verifiedrag query-rewriting hyde multi-query step-back-prompting retrieval information-retrieval vector-search
Fix the most common RAG failure — bad retrieval because the user's raw question doesn't match any chunk — with HyDE, multi-query, step-back prompting, and query decomposition.
-
test-time-compute-scaling verifiedtest-time-compute reasoning inference best-of-n self-consistency mcts prm verifier llm
Scale reasoning quality at inference — chain-of-thought, self-consistency, best-of-N, verifier/PRM-guided search, and MCTS — with the sample-efficiency and cost tradeoffs of each.
-
attention-kernel-optimization verifiedattention flashattention paged-attention kv-cache gpu kernel inference long-context llm optimization
Understand and apply the GPU-kernel optimizations that make long-context inference possible — FlashAttention 1/2/3, PagedAttention, Flash-Decoding, and FlashInfer — with when to reach for each.
-
agent-skills-progressive-disclosure verifiedagents skills progressive-disclosure skilly-md claude context tooling anthropic
Design and package Agent Skills — Anthropic's file-and-folder pattern (SKILL.md + progressive disclosure) for giving agents procedural knowledge, now a cross-platform open standard.
-
grpo-reasoning-finetune verifiedgrpo rlvr reinforcement-learning reasoning r1 fine-tuning trl reward-function
Fine-tune a base model to reason with Reinforcement Learning from Verifiable Rewards (RLVR) using GRPO — reward function design, group sampling, KL control, and when RL beats SFT for math/code/agentic tasks.
-
mcp-client-development verifiedmcp model-context-protocol client python sdk llm tool-use agents
Build an MCP client — connect to stdio and streamable-HTTP servers, run the initialize handshake, list tools/resources, call tools, handle errors, and wire tools into an LLM agent loop.
-
mcp-server-development verifiedmcp model-context-protocol fastmcp tools agents api sdk
Build and ship a Model Context Protocol (MCP) server that agents can call — tools, resources, prompts, stdio vs streamable-HTTP transports, lifecycle, and testing.