llm-guard-kit
v0.17.0: MiniJudge ($0 AUROC 0.747, distilled from Sonnet), cross-domain TV validated AUROC 0.660 [CI 0.614-0.705] n=1000, QuickCalibrator, 4 platform integrations (Langfuse/LangSmith/Prometheus/Datadog). Real-time reliability monitoring for LLM agents. v0.16.1: fix FastAPI>=0.115 (starlette 0.52 compat), domain-invariant feature selection 0.778 cross-domain. v0.16.0: hosted calibration endpoint, multilevel features, 3-domain validation, latency SLA. v0.15.0: probe_ensemble_blend() +1.6pp AUROC.
Directory Presence
Cross-referenced across 19 tracked directories
Adoption Metrics
#109
Popularity Rank
5%
Adoption Rate
Emerging
Adoption Stage
18
Unlisted Directories
Recently added to directories
Cross-Posting Gaps
Not yet listed in these active directories:
Related Agents
UI TARS Desktop
bytedance
The Open-Source Multimodal AI Agent Stack: Connecting Cutting-Edge AI Models and Agent Infra
E2B
e2b-dev
Open-source, secure environment with real-world tools for enterprise-grade agents.
X Twitter Scraper
Xquik-dev
X (Twitter) automation skill for AI coding agents. 60+ API endpoints, 20 MCP tools. Tweet search, user lookup, follower extraction, write actions (tweet, like, retweet, follow, DM), media download, account monitoring & trending topics. REST API, MCP server, HMAC webhooks. Works with Claude Code, Cursor, Codex, Copilot, Windsurf & 40+ agents.
Instar
SageMindAI
Persistent Claude Code agents with scheduling, sessions, memory, and Telegram.