Search

LinkedIn content management built as a Claude Code skill. Research topics, generate posts in your voice, source and brand images via Canva, review drafts and publish on schedule all from your terminal.

A set of Claude skills I built for myself as a neurodivergent professional. If your brain works anything like mine, they might help you too.

Founder OS skills for Claude Code -- executive function, voice, debrief, session management, and delivery validation tools.

No description available

轻量级 LLM 应用可观测性 SDK — 调用追踪、提示词工程、批量处理

Scan your Elasticsearch/OpenSearch cluster. Find dead indices, wasted storage, and cost savings.

Oasis LLM Proxy Client

Alliance Auth plugin: PI Manager (FYAD) – track PI installations, monitor storage and extractors, share planet templates.

OrbiGPU — Unified OpenAI-compatible API gateway for 100+ LLM models. Official SDK coming in v0.1.0.

Domain-agnostic Graph RAG framework for building governed, auditable Knowledge Graphs

Persistent memory for Claude Code — knowledge graph in Obsidian vault

Add Claude as middleware to any Python app with a single decorator

Structured document processor with diagram/image/text extraction with optional langchain or dataset output

pytest plugin for semantic LLM output testing — validate meaning, not just shape.

GPU-Accelerated RDF2Vec – A high-performance GPU implementation of RDF2Vec that uses CUDA and RAPIDS to generate scalable, embeddings for large dense knowledge graphs.

Search AI podcast quotes by topic — recent episode excerpts from Lex Fridman, Dwarkesh Patel, No Priors, Latent Space, and dozens more. Universal agentskills.io skill compatible with Claude Code, OpenAI Codex, Hermes Agent, OpenClaw, and any other agent t

Cloudflare Wrangler multi-account manager — save, switch, and run wrangler against multiple Cloudflare Workers accounts with AWS-style --profile and per-invocation shadow HOME isolation.

CLI for installing shared AI commands, rules, and skills into Copilot and Cursor.

API Security Testing Plugin for OpenCode - Automated vulnerability scanning and penetration testing

Snapshot tests for LLM tool calls — record once, replay deterministically in CI. Catch regressions in your agent wiring before they hit production.