Harness Bridge
SDK / ToolA local launcher that points Claude Code and seven other coding harnesses at any OpenAI-, Anthropic- or Responses-compatible endpoint — without editing a single config file.
- launcher
- local-models
- cli
- menu-bar
AgentOps
Observability for AI agents — session replay, LLM cost tracking, and monitoring in a few lines.
AgentOps is a DevTool platform for AI agents: drop-in SDK instrumentation that records session timelines (every LLM call, tool use, and cost) and gives you replay, alerting, and evaluation dashboards. The pragmatic choice when an agent built on Claude or any other model starts misbehaving in production and logs alone don't explain why.
What it does
- Agent observability SDK: session replays, LLM call analytics, cost tracking in 2 lines
- Auto-instruments OpenAI Agents SDK, LangChain, CrewAI and other frameworks
- Dashboard at app.agentops.ai for traces, evals and debugging
When to use it
When an agent misbehaves in production: replay the session, see every LLM call with latency/cost, instead of adding print statements.
Getting started
pip install agentops
Then in code:
import agentops
agentops.init('<API_KEY>')
Get a key at app.agentops.ai/settings/projects.Things to know
Sends session data to a hosted SaaS — prompts and tool outputs leave your infra; check data policies for sensitive workloads. Free tier limits apply.
Repository
AgentOps-AI/agentopsCharted
Related on the map
Claude for Siri
SDK / ToolAnswers macOS Spotlight and Siri prompts with your signed-in Claude Code account — an experimental macOS 27 model-delegation provider.
- macos
- siri
- spotlight
- model-delegation
AI Coding Assistant Data Extractor
SDK / ToolExtract your own local chat history from ten AI coding assistants into one normalized JSONL — for fine-tuning, analytics, or backing up years of conversations. Standard library only.
- cli
- data-export
- jsonl
- backup