llm-tooling

Better AI Predictions ⚡

Artifact-first agent framework — disciplined skill chain for AI-driven multi-step engineering tasks. MIT.

AI agent skills library built by OpenSite AI. A single git repo that keeps Claude, Codex, Cursor, Copilot, and Perplexity Computer in perfect skill sync, with persistent cross-session memory and automated cloud upload via Playwright.

An orientation protocol for LLM sessions — plus a curated Claude skills collection

Template-aware development workflow plugin for Claude Code.

Stop fighting broken PDF tool calls. This watches a folder and converts any PDF to clean Markdown in seconds — fully local, no API key, no rate limits, no cloud dependency. Built for Obsidian "second brain" setups paired with Claude or any LLM assistant with filesystem access. Free forever.

Per-repo memory, outcome telemetry, and a calibrated-confidence gate for Claude Code, with MCP and AGENTS.md projections so other AI coding tools can read its context. Notes survive sessions; success claims need test evidence; your reverts are remembered. Local-only, stdlib runtime.

finasys: Fast financial data processing, feature engineering & AI agent toolkit

System-level lint for multi-agent harnesses. Catches the 21 structural traps single-file linters miss — including the LLM-when-you-should-use-code patterns that burn tokens.

Give Claude structured decisions: one label from your allowed set with explicit confidence, first-class abstention, and auditable evidence, plus a policy gate that decides when automation is allowed.

Fast BPE tokenizer with guaranteed perfect text reconstruction. 2x faster than tiktoken at encoding/decoding while preserving all whitespace and formatting.

🤖 operational-ai-agents · ⚡ 13 agentes para Claude Code · 🧭 repository-evolution · 🏗️ legacy-modernization · 📡 curriculum-evolution · 🌐 portfolio-publication · 🛡️ security-remediation · 🩺 incident-root-cause · 👤 professional-profile · Contrato + gates humanos + evidencia verificable · 39 evals deterministas · Zero-deps · Cross-platform 🐧🍎🪟

A hook-based router for Claude Code: classifies every prompt, dispatches to the right framework, and enforces runtime guardrails.

Local real-time coaching for League of Legends and Teamfight Tactics. A deterministic engine does the item and damage math from the game's own numbers, and the AI coach only narrates what that math already decided. No win-rate scraping.

A practical workbench of tools + methodology for solo devs driving an AI coding agent (Claude Code, Cursor, Copilot): safety hooks, secrets crypto, invariant + test-selection tooling, leak scanner, plus skill & memory system templates.

The primitives layer for every coding agent — one handler API over per-agent hooks, instruction files, plugin packaging, and config.

Context-rot monitor for Claude Code: statusline meter grades your context window SHARP/OK/FADING/DUMB, and auto-writes a session handoff at 90%.

Create evidence-bound AI-agent contracts, governed runtime flows, assurance reports, trace/evidence bundles, and falsification-aware IX-CognitionKernel Wave 6 cognition-contract handoff packages. No AGI claim.

Ferramenta repo-to-text para LLMs. Transforma toda a estrutura e o conteúdo de um projeto em um único digest de texto filtrado e pronto para o prompt da sua IA.

One dispatcher, many runners, one problem. A local control plane for agent CLI execution.