A curated set of skills
This repository also configures its own agents. See what skill-quiver tells them →
Skill Claude CodeCodex
Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.
Skill Claude CodeCodex
Serverless GPU cloud platform for running ML workloads. Use when you need on-demand GPU access without infrastructure management, deploying ML models as APIs, or running batch jobs with automatic scaling.
Skill Claude CodeCodex
OBLITERATUS: abliterate LLM refusals (diff-in-means).
Skill Claude CodeCodex
Guarantee valid JSON/XML/code structure during generation, use Pydantic models for type-safe outputs, support local models (Transformers, vLLM), and maximize inference speed with Outlines - dottxt.ai's structured generation library.
Skill Claude CodeCodex
Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.
Skill Claude CodeCodex
Expert guidance for Fully Sharded Data Parallel training with PyTorch FSDP - parameter sharding, mixed precision, CPU offloading, FSDP2.
Skill Claude CodeCodex
SAM: zero-shot image segmentation via points, boxes, masks.
Skill Claude CodeCodex
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.
Skill Claude CodeCodex
Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.
Skill Claude CodeCodex
Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization.
Skill Claude CodeCodex
Use when deploying production LLM APIs, optimizing inference latency/throughput, or serving models with limited GPU memory. Supports OpenAI-compatible endpoints, quantization (GPTQ/AWQ/FP8), and tensor parallelism.
Skill Claude CodeCodex
W&B: log ML experiments, sweeps, model registry, dashboards.
Skill Claude CodeCodex
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…
Skill Claude CodeCodex
Orchestrate multiple AI agents on the same host. Two modes: dispatch (one-shot tasks routed to specialist agents via CLI runners) and peer agent management (persistent processes with file-based inbox/outbox communication). Includes zombie hygiene, process monitoring, and parallel dispatch patterns. TRIGGER when: user…
Skill Claude CodeCodex
Suggests manual context compaction at logical intervals to preserve context through task phases rather than arbitrary auto-compaction.
Skill Claude CodeCodex
Cost optimization patterns for LLM API usage — model routing by task complexity, budget tracking, retry logic, and prompt caching.
Skill Claude CodeCodex
KB chore runner — organizes unprocessed notes, enforces structure, rotates live entries, tag hygiene. Surfaces findings for human review.
Skill Claude CodeCodex
OpenCode CLI reference — delegate coding tasks to OpenCode as an autonomous coding worker. Covers one-shot tasks, interactive background sessions, PR review, parallel work patterns, and session management. Requires OpenCode installed and auth configured.
Skill Claude CodeCodex
Create, deploy, and manage long-running resilient background service scripts that persist across agent and system restarts. Covers PID file management, logging, systemd deployment, and verification. TRIGGER when: user asks to "run a script in background", "keep running after restart", "create a daemon", "background…
Skill Claude CodeCodex
Bidirectional retrospective — grades agent system evolution and human decision quality. Produces a structured review note in your KB.
Skill Claude CodeCodex
Distills a Claude Code session into a structured KB note following the above/below the line format. Captures actionable insights, not raw transcripts.
Skill Claude CodeCodex
Execute implementation plans by dispatching fresh subagents per task with two-stage review (spec compliance then code quality). No agent verifies its own work. Fresh context per task prevents state pollution. TRIGGER when: you have an implementation plan ready and want to execute it task-by-task with systematic…
Skill Claude CodeCodex
Manage system-level cron jobs for the default user using the crontab command. Use for adding, listing, and removing scheduled tasks that are managed by the OS, not the agent's internal cron tool.
Skill Claude CodeCodex
Systematically audit a Linux VPS — find all listening services, check firewall rules (UFW + iptables), test external reachability, inspect logs for blocked probes. Answer the question: "what's exposed to the internet?".
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: