Skills

n8n node to read and control Flagsmith feature flags, and trigger workflows on flag changes

n8n community node for Qorami — check an email before your workflow sends it (send / ask a human / block).

Python SDK for observability, tracing, and evaluation of LLM applications

Memory tools for the llm-chat ecosystem

LLM-as-judge linter: enforce code-quality checks deterministic linters can't express, by driving real coding harnesses through oneharness.

Ragmir add-on for plug-and-play Edge-quality and offline text-to-speech audio summaries.

A local-first, config-driven LLM router across OpenAI-compatible backends you own.

Persistent markdown knowledge base (OKF format) for LLM project memory. Two-tier wiki with immutable raw sources and LLM-generated concept pages — consult it before answering questions about project facts, architecture, design decisions, or domain entitie

A super basic Pi extension that toggles GPT-5.5 Codex Fast mode with /fast (the default 'priority'), nothing else - one file, 180 lines of code.

Makes a second request to the LLM, asking it if its original response was relevant to the prompt.

LangChain/LangGraph callback adapter feeding graph-trace.

Publication-quality LLM architecture diagrams, fact sheets, and diffs from HuggingFace config.json — no weights needed

Finds, scores, and ranks the best library, SDK, or tool for your exact stack — no Reddit threads, no outdated blogs. 5-tier search, deterministic scoring, built-in security gate. Works with Claude Code, Claude.ai, Codex & Antigravity.

Extract skills from job postings and resumes — 32K-skill gazetteer + MiniLM embeddings + MLP context classifier (73% F1, trained on 491K samples). ONNX Runtime, no torch.