nix-flake-dotfiles-maintenance
Use when modifying a multi-host Nix flake or dotfiles repo, especially host-specific inputs, overlays, lockfile scope, and evaluation checks.
kalshi-backtest-artifact-controls
Mandatory artifact controls when assaying Kalshi (or any kline+tape) backtest signals: bar-label look-ahead audit, strict trade-through maker fills, month-split stability, adverse-selection fill checks
maps
Geocode, POIs, routes, timezones via OpenStreetMap/OSRM.
kalshi-delta-replay-dislocation-timing
Measure sub-second cross-market dislocation lifetimes on Kalshi lake data (any family) via validated orderbook_delta replay — use for arb-resolution timing, ladder coherence, or two-leg re-sync questions in kairos-research.
kde-mixture-experiment-forensics
Forensic checklist for reviewing local/matchup-level KDE or kernel-mixture density experiments (SEAM-style, conditional spray/density heads) whose local model loses to a pooled marginal baseline or shows a null conditioning effect
himalaya
Himalaya CLI: IMAP/SMTP email from terminal.
minecraft-modpack-server
Host modded Minecraft servers (CurseForge, Modrinth).
kairos-hotswap-disk-transfer-leg
Run a Kairos migration disk-to-disk copy leg on emc after a drive (PD1/PD2) is inserted into the hot-swap bay — identity gating by UUID, writer checks, persistent systemd rsync units. Use when the user says a PD drive was connected to emc and asks to initiate/continue the transfer.
kairos-emc-disk-growth-assessment
Diagnose rapid emc-local disk growth without destructive scans or changes
pokemon-player
Play Pokemon via headless emulator + RAM reads.
codebase-inspection
Inspect codebases w/ pygount: LOC, languages, ratios.
grpo-rl-training
Expert guidance for GRPO/RL fine-tuning with TRL for reasoning and task-specific model training
arxiv
Search arXiv papers by keyword, author, category, or ID.
kairos-crypto-hunt-safety
Use when running parallel Kalshi crypto lake assays in kairos-research; enforces safe DuckDB resource limits, causal signal controls, fill realism, and verified fee/L2 decoding.
get-pr-comments
Retrieve and filter GitHub pull request comments with the gh CLI.
notion
Notion API + ntn CLI: pages, databases, markdown, Workers.
kairos-collector-performance-baseline
Capture and interpret a durable before/after performance baseline for Kairos collector workers and orchestrator
webhook-subscriptions
Webhook subscriptions: event-driven agent runs.
offline-vap-cel-verification
Verify or safely change a Kubernetes ValidatingAdmissionPolicy (CEL) offline when the live admission gate is unavailable — golden-pin exact expressions, generate fixtures from the real renderer, run an independent intent-evaluator, and get an independent review.
kanban-worker
Pitfalls, examples, and edge cases for Hermes Kanban workers. The lifecycle itself is auto-injected into every worker's system prompt as KANBAN_GUIDANCE (from agent/prompt_builder.py); this skill is what you load when you want deeper detail on specific scenarios.
kanban-orchestrator
Decomposition playbook + anti-temptation rules for an orchestrator profile routing work through Kanban. The "don't do the work yourself" rule and the basic lifecycle are auto-injected into every kanban worker's system prompt; this skill is the deeper playbook when you're specifically playing the orchestrator role.
petdex
Install and select animated petdex mascots for Hermes.
pytorch-lightning
High-level PyTorch framework with Trainer class, automatic distributed training (DDP/FSDP/DeepSpeed), callbacks system, and minimal boilerplate. Scales from laptop to supercomputer with same code. Use when you want clean training loops with built-in best practices.
nano-pdf
Edit PDF text/typos/titles via nano-pdf CLI (NL prompts).
powerpoint
Create, read, edit .pptx decks, slides, notes, templates.
spotify
Spotify: play, search, queue, manage playlists and devices.
github-auth
GitHub auth setup: HTTPS tokens, SSH keys, gh CLI login.
github-code-review
Review PRs: diffs, inline comments via gh or REST.
requesting-code-review
Use when completing tasks, implementing major features, or before merging to verify work meets requirements
fine-tuning-with-trl
TRL: SFT, DPO, PPO, GRPO, reward modeling for LLM RLHF.
arch-aur-package-management
Use when setting up, repairing, or using Arch User Repository (AUR) workflows on Arch Linux: prerequisites, yay/paru helper installation, PKGBUILD downloads, libalpm rebuilds after pacman upgrades, and verification.
distributed-llm-pretraining-torchtitan
Provides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when pretraining Llama 3.1, DeepSeek V3, or custom models at scale from 8 to 512+ GPUs with Float8, torch.compile, and distributed checkpointing.
godmode
Jailbreak LLMs: Parseltongue, GODMODE, ULTRAPLINIAN.
simpo-training
Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4 points on AlpacaEval 2.0). No reference model needed, more efficient than DPO. Use for preference alignment when want simpler, faster training than DPO/PPO.
slime-rl-training
Provides guidance for LLM post-training with RL using slime, a Megatron+SGLang framework. Use when training GLM models, implementing custom data generation workflows, or needing tight Megatron-LM integration for RL scaling.
using-proton-pass-cli
Interact with Proton Pass via the `pass-cli` command-line tool. Use this skill when the user wants to manage passwords, vaults, items, SSH keys, secret injection, or any credential management through the Proton Pass CLI. Triggers on mentions of pass-cli, proton pass, secret references, pass:// URIs, vault management, or SSH agent integration with Proton Pass.
axolotl
Axolotl: YAML LLM fine-tuning (LoRA, DPO, GRPO).
dspy
DSPy: declarative LM programs, auto-optimize prompts, RAG.
whisper
OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M params). Use for speech-to-text, podcast transcription, or multilingual audio processing. Best for robust, multilingual ASR.
unsloth
Unsloth: 2-5x faster LoRA/QLoRA fine-tuning, less VRAM.
chroma
Open-source embedding database for AI applications. Store embeddings and metadata, perform vector and full-text search, filter by metadata. Simple 4-function API. Scales from notebooks to production clusters. Use for semantic search, RAG applications, or document retrieval. Best for local development and open-source projects.
faiss
Facebook's library for efficient similarity search and clustering of dense vectors. Supports billions of vectors, GPU acceleration, and various index types (Flat, IVF, HNSW). Use for fast k-NN search, large-scale vector retrieval, or when you need pure similarity search without metadata. Best for high-performance applications.
stable-diffusion-image-generation
State-of-the-art text-to-image generation with Stable Diffusion models via HuggingFace Diffusers. Use when generating images from text prompts, performing image-to-image translation, inpainting, or building custom diffusion pipelines.
segment-anything-model
SAM: zero-shot image segmentation via points, boxes, masks.
stacked-prs
Push local branches and create stacked PRs. Analyzes branch topology, pushes, creates PRs with correct base branches, and resolves merge conflicts. Use when you have a chain of local branches to turn into stacked PRs.
stacked-issues
Implement multiple GitHub issues sequentially as stacked branches in separate worktrees, with an implementer sub-agent and an independent reviewer sub-agent per issue. Use when the user gives you two or more dependent issues and asks for them to be implemented in order, or says "stacked branches", "sequential issues", "issue chain", "do these in worktrees", or describes a parent epic with child issues that build on each other. Also reach for this whenever the user wants implementation and verification done by separate agents.
github-issues
Create, triage, label, assign GitHub issues via gh or REST.
github-pr-workflow
GitHub PR lifecycle: branch, commit, open, CI, merge.
using-nautilus-trader
Use when building, backtesting, or deploying algorithmic trading systems with NautilusTrader (nautilus_trader) — writing Strategy/Actor classes and StrategyConfig, configuring BacktestNode/BacktestEngine, constructing instruments, order management, ParquetDataCatalog/wranglers, message bus and custom data streams, indicators, portfolio/analysis/report generation, or live TradingNode deployment. Curated for v1.230.0.
pinecone
Managed vector database for production AI applications. Fully managed, auto-scaling, with hybrid search (dense + sparse), metadata filtering, and namespaces. Low latency (<100ms p95). Use for production RAG, recommendation systems, or semantic search at scale. Best for serverless, managed infrastructure.
Page 1541 of 1700 · 84978 results
