LLMOps · Eval

Raven wraps agents in durable memory and self-refining skills so workflows survive past the chat session.

Token Monitor reads local logs from two dozen AI coding tools to surface live token burn, costs, and limits in one place, synced across all your machines.

A hybrid CLI tool that uses deterministic pipelines to keep LLM agents from drifting off-target during code review.

Most AI scientist tools are monolithic prompt pipelines; FAROS treats research automation as a composable runtime problem rather than a single-agent stack.

It exists to stop your AI gateway from quietly burning through quotas, cash, and expired OAuth tokens without leaving a paper trail.

To wire up public A-share, US, and HK market data into a single local dashboard and let your own AI analyze it without pretending to pick winners.

It replaces flat vector dumps with a four-tier semantic pyramid and Mermaid symbol graphs so agents remember workflows without drowning in their own tool logs.

Books are too good to leave on the shelf; this system distills them into structured, callable agent skills.

A Python layer that makes AI agents explain themselves through structured context graphs, decision trails, and W3C-compliant provenance.

repowise indexes a codebase into five queryable intelligence layers—dependency graphs, git history, docs, architectural decisions, and deterministic health scores—so MCP-compatible agents can answer "why" instead of grepping for "what".

It chains a dozen LLM agents into a visual workflow so your on-call engineer only has to tap 'approve' instead of SSHing in at 3 AM.

An open-source audit engine that scores how likely ChatGPT, Perplexity, and Gemini are to cite your site—and generates the fixes.

A cross-CLI skill that turns your static notes into a self-updating knowledge base for Claude, Codex, Gemini, and OpenCode.

DeepSWE measures whether frontier coding agents can complete real, long-horizon engineering tasks from active open-source repositories—not just generate snippets, but ship verifiable patches.

Chinese AI output too often reads like a blend of corporate press release and therapy session; this skill file teaches models to cut the performance and keep the facts.

It packages a professor's decade of SIGMOD and NeurIPS experience into structured AI skills, bridging the 'last-mile' gap where generic guides and busy advisors leave grad students stranded.

To keep scientific AI local: a desktop workbench where LLMs run Python, R, and query ~80 bio DBs without cloud lock-in.

Reusable instruction modules that give AI agents structured playbooks for repeatable coding, research, and ops tasks.

iFixAi runs up to 32 inspections against any LLM or agent and returns a letter-grade scorecard in minutes, using a separate provider as judge so the model isn't grading its own homework.

VulnClaw exists so that a single natural-language sentence can trigger the entire reconnaissance-to-report pipeline without manually orchestrating a dozen separate tools.