Agent Ops Python Packages

neuralbridge-sdk

NeuralBridge - AI Agent 自愈引擎。三层故障自愈：L1 自动重试 → L2 语义等价降级 → L3 Failover 兜底。内置本地监控面板，依赖 httpx + aiohttp，pip install 即可接入。

agent-discover-scanner

The industry-standard Agentic Identity & Inventory Scanner. Automatically inventory autonomous agents (LangChain, AutoGen, CrewAI, PydanticAI) using static analysis, network heuristics, and eBPF. Foundational tool for AIBOM compliance and AgentOps governance.

agentdiscover

The industry-standard Agentic Identity & Inventory Scanner. Automatically inventory autonomous agents (LangChain, AutoGen, CrewAI, PydanticAI) using static analysis, network heuristics, and eBPF. Foundational tool for AIBOM compliance and AgentOps governance.

helm-agent-ops

Stability-first operations CLI for long-running AI agent workspaces.

openclaw-upgrade-orchestrator-mcp

MCP server for safe AI agent runtime upgrades — version-aware regression catalog, pre/post snapshot diff, rollback guides. Read-only advisor: never executes upgrades. Bring your own catalog for non-OpenClaw runtimes.

silentwatch-mcp

MCP server for catching cron silent failures in production AI deployments — exit-0-with-empty-output detection, retry storms, action-budget leaks. Works with system cron, systemd timers, OpenClaw cron logs.

openclaw-cost-tracker-mcp

MCP server for AI agent token-cost telemetry across Anthropic, OpenAI, Gemini, Ollama — per-agent attribution, anomaly detection, model-routing recs with savings estimates, monthly forecast.

openclaw-health-mcp

MCP server for AI agent deployment health — gateway, CPU/RAM, recent errors, skill registry, upgrades, disk usage. Cross-platform Linux-proc backend; OpenClaw operators get native ~/.openclaw/ parsing.

replayd

Turn failed AI agent runs into replayable regression tests. Catch regressions before you ship.

liteswarm

Lightweight Agent Framework for building AI apps with any LLM