semantica-agi/semantica
Semantica is an AI-native knowledge graph intelligence framework that provides an auditable, governed, and explainable context and accountability layer for AI agents and LLM systems.
Awesome Infra for AI › AI Governance & Compliance
Floe-Labs/floe-guard
floe-guard enforces a monetary ceiling on AI agent workloads at the moment of the call rather than reporting the overspend afterwards. It runs in process, needs no account or signup, and emits no telemetry by default. The core interface is a pair of check and record calls: the guard is asked whether the next unit of paid work fits under the remaining budget, and the cost of completed work is recorded back. Around that core sit adapters that wire the check automatically into common runtimes, including OpenAI, Anthropic and Gemini clients, CrewAI, LiteLLM, LangChain and LangGraph on the Python side, and the Vercel AI SDK on the TypeScript side. Voice agents get first-class treatment: adapters for Pipecat, LiveKit, Vapi and Retell meter speech-to-text, text-to-speech, model and telephony cost per call, and pre-call admission gates can refuse a call before it is answered. Paid tools are covered by a reserve and settle pair that blocks before the call runs, in contrast to recording after the fact, which can account for spend but cannot prevent it. The practical effect is that a runaway loop terminates at a small configured ceiling instead of accumulating a large bill. Both a Python package and an npm package are published, with an adapter matrix documenting which integrations exist in each. The repository also self-describes for coding agents through machine-readable onboarding files so an assistant can install and wire the guard unattended. It fits teams running agents or voice pipelines where per-session cost is variable and a hard limit is a requirement rather than a report.
https://github.com/Floe-Labs/floe-guard
Also in AI Governance & Compliance
iflytek/skillhub
SkillHub is a self-hosted, open-source agent skill registry for enterprises to publish, version, discover, and govern reusable skill packages for AI agents with RBAC and audit logs.
kitops-ml/kitops
KitOps is a CNCF open-source tool for packaging, versioning, and securely sharing AI/ML models, datasets, code, and configuration into an OCI Artifact for simplified deployment and governance.
verifywise-ai/verifywise
VerifyWise is an AI governance platform providing tools for LLM evaluation, risk management, and compliance with AI regulations including the EU AI Act, ISO 42001, and NIST AI RMF.