197 tools found
jevals runs cheap, fast agent evals and guardrails using Jev decision models instead of an LLM judge.
Self-hosted bring-your-own-key AI workspace: chat with 300+ models, build self-drafting agents, run research and code-gen, keys sealed in the browser.
Turns one goal into an editable plan of coding-agent tasks, each with its own runner and model, then runs and verifies them by completion marker.
agent-dev-team routes coding work across 21 tiered role agents with a structured escalation protocol, so no single assistant attempts everything.
A local-first SQLite episodic memory engine that keeps decisions and state intact across Claude Code, Cursor, Codex, and other agents.
Framework-free Colab notebooks for AI Engineer and FDE skills: model APIs, RAG, evals, agents, security, serving and case studies on free Groq.
Open source AI productivity OS on Cloudflare Workers: agent chat, sandboxed gadget apps and Gatekeepers that add capability-based guardrails.
Multiplayer agent harness for startups in Slack and on the web, with isolated per-person workspaces, shared rooms, skills and admin security postures.
Model-neutral MCP coding runtime: file edits, patches, command execution, and git for any MCP client.
Real-time detection, instance, and semantic segmentation in one framework and config flag, with multi-backend export (TensorRT, OpenVINO, CoreML).
Open-source platform giving every employee a sandboxed AI agent, routed through a credential-injecting gateway with one team-wide policy.
Open-source, hash-chained audit trail for AI agents - tamper-evident Runtime Records rendered as a self-verifying HTML report customers can check.
