graniet/kheish

Orchestration engine for long-running, tool-using AI agents

This repository also configures its own agents. See what kheish tells them →

227Stars on the repository
79Mods indexed here, across every type
1mo agoLast push, which is what freshness is scored on
Apache-2.0Licence, which decides whether bodies are shown

dspy

49

graniet/kheish

Skill Claude CodeCodex

Build complex AI systems with declarative programming, optimize prompts automatically, create modular RAG systems and agents with DSPy - Stanford NLP's framework for systematic LM programming.

not rated 227 1mo ago A 32 tokens copy · 80% Apache-2.0

axolotl

50

graniet/kheish

Skill Claude CodeCodex needs its repo

Expert guidance for fine-tuning LLMs with Axolotl - YAML configs, 100+ models, LoRA/QLoRA, DPO/KTO/ORPO/GRPO, multimodal support.

not rated 227 1mo ago A 43 tokens copy · 81% Apache-2.0

grpo-rl-training

51

graniet/kheish

Skill Claude CodeCodex

Expert guidance for GRPO/RL fine-tuning with TRL for reasoning and task-specific model training.

not rated 227 1mo ago A 20 tokens copy · 95% Apache-2.0

peft

52

graniet/kheish

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.

not rated 227 1mo ago A 75 tokens copy · 80% Apache-2.0

pytorch-fsdp

53

graniet/kheish

Skill Claude CodeCodex

Expert guidance for Fully Sharded Data Parallel training with PyTorch FSDP - parameter sharding, mixed precision, CPU offloading, FSDP2.

not rated 227 1mo ago B 31 tokens copy · 98% Apache-2.0

trl-fine-tuning

54

graniet/kheish

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.

not rated 227 1mo ago A 62 tokens copy · 72% Apache-2.0

unsloth

55

graniet/kheish

Skill Claude CodeCodex

Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization.

not rated 227 1mo ago A 35 tokens copy · 86% Apache-2.0

obsidian

56

graniet/kheish

Skill Claude CodeCodex

Read, search, and create notes in the Obsidian vault.

not rated 227 1mo ago A SkillSpector: warn 14 tokens original Apache-2.0

google-workspace

57

graniet/kheish

Skill Claude CodeCodex

Gmail, Calendar, Drive, Contacts, Sheets, and Docs integration via gws CLI (googleworkspace/cli). Uses OAuth2 with automatic token refresh via bridge script. Requires gws binary.

not rated 227 1mo ago A SkillSpector: warn 40 tokens original Apache-2.0

linear

58

graniet/kheish

Skill Claude CodeCodex

Manage Linear issues, projects, and teams via the GraphQL API. Create, update, search, and organize issues. Uses API key auth (no OAuth needed). All operations via curl — no dependencies.

not rated 227 1mo ago A 41 tokens copy · 81% Apache-2.0

nano-pdf

59

graniet/kheish

Skill Claude CodeCodex

Edit PDFs with natural-language instructions using the nano-pdf CLI. Modify text, fix typos, update titles, and make content changes to specific pages without manual editing.

not rated 227 1mo ago A SkillSpector: pass 34 tokens original Apache-2.0

notion

60

graniet/kheish

Skill Claude CodeCodex

Notion API for creating and managing pages, databases, and blocks via curl. Search, create, update, and query Notion workspaces directly from the terminal.

not rated 227 1mo ago A SkillSpector: warn 33 tokens original Apache-2.0

ocr-and-documents

61

graniet/kheish

Skill Claude CodeCodex

Extract text from PDFs and scanned documents. Use webextract for remote URLs, pymupdf for local text-based PDFs, marker-pdf for OCR/scanned docs. For DOCX use python-docx, for PPTX see the powerpoint skill.

not rated 227 1mo ago A 49 tokens copy · 84% Apache-2.0

powerpoint

62

graniet/kheish

Skill Claude CodeCodex

Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying…

not rated 227 1mo ago A 149 tokens copy · 83% Apache-2.0

godmode

63

graniet/kheish

Skill Claude CodeCodex ⚠ unsafe

Jailbreak API-served LLMs using G0DM0D3 techniques — Parseltongue input obfuscation (33 techniques), GODMODE CLASSIC system prompt templates, ULTRAPLINIAN multi-model racing, encoding escalation, and Kheish-native prefill/system prompt integration. Use when a user wants to bypass safety filters, jailbreak a model, or…

not rated 227 1mo ago F ⚑ AI: unsafe 81 tokens copy · 84% Apache-2.0

arxiv

64

graniet/kheish

Skill Claude CodeCodex

Search and retrieve academic papers from arXiv using their free REST API. No API key needed. Search by keyword, author, category, or ID. Combine with webextract or the ocr-and-documents skill to read full paper content.

not rated 227 1mo ago C 49 tokens copy · 94% Apache-2.0

blogwatcher

65

graniet/kheish

Skill Claude CodeCodex

Monitor blogs and RSS/Atom feeds for updates using the blogwatcher-cli tool. Add blogs, scan for new articles, track read status, and filter by category.

not rated 227 1mo ago A 34 tokens copy · 91% Apache-2.0

llm-wiki

66

graniet/kheish

Skill Claude CodeCodex

Karpathy's LLM Wiki — build and maintain a persistent, interlinked markdown knowledge base. Ingest sources, query compiled knowledge, and lint for consistency.

not rated 227 1mo ago B 32 tokens copy · 81% Apache-2.0

polymarket

67

graniet/kheish

Skill Claude CodeCodex

Query Polymarket prediction market data — search markets, get prices, orderbooks, and price history. Read-only via public REST APIs, no API key needed.

not rated 227 1mo ago A 33 tokens copy · 84% Apache-2.0

graniet/kheish

Skill Claude CodeCodex

End-to-end pipeline for writing ML/AI research papers — from experiment design through analysis, drafting, revision, and submission. Covers NeurIPS, ICML, ICLR, ACL, AAAI, COLM. Integrates automated experiment monitoring, statistical analysis, iterative writing, and citation verification.

not rated 227 1mo ago B 60 tokens copy · 84% Apache-2.0

openhue

69

graniet/kheish

Skill Claude CodeCodex

Control Philips Hue lights, rooms, and scenes via the OpenHue CLI. Turn lights on/off, adjust brightness, color, color temperature, and activate scenes.

not rated 227 1mo ago A SkillSpector: pass 32 tokens original Apache-2.0

xitter

70

graniet/kheish

Skill Claude CodeCodex

Interact with X/Twitter via the x-cli terminal client using official X API credentials. Use for posting, reading timelines, searching tweets, liking, retweeting, bookmarks, mentions, and user lookups.

not rated 227 1mo ago B 42 tokens copy · 89% Apache-2.0

plan

71

graniet/kheish

Skill Claude CodeCodex

Plan mode for Kheish — inspect context, write a markdown plan into the active workspace's .kheish/plans/ directory, and do not execute the work.

not rated 227 1mo ago A SkillSpector: pass 36 tokens original Apache-2.0

graniet/kheish

Skill Claude CodeCodex

Pre-commit verification pipeline — static security scan, baseline-aware quality gates, independent reviewer subagent, and auto-fix loop. Use after code changes and before committing, pushing, or opening a PR.

not rated 227 1mo ago A SkillSpector: warn 41 tokens original Apache-2.0

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: