From ai-telemetry
Design a practical AI telemetry or OpenTelemetry setup for Claude, ChatGPT, Codex, internal agents, or multi-tool AI programs. Use when the user asks how to turn on AI logging, connect AI tools to Honeycomb or another OTEL backend, decide what to log or redact, create a first telemetry rollout plan, or point an AI assistant at vendor monitoring docs and ask what makes sense for an organization.
How this skill is triggered — by the user, by Claude, or both
Slash command
/ai-telemetry:ai-otel-setup-advisorThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Help the user turn AI telemetry on in a way that answers real management questions without over-collecting sensitive data.
Help the user turn AI telemetry on in a way that answers real management questions without over-collecting sensitive data.
Identify the target setup:
If exact vendor configuration matters, check current official docs before giving commands. Defaults and environment variables change.
Start with observe-only telemetry:
Design around the five metric families:
Use a single session_id or trace identifier wherever possible so work can be followed across tools.
Do not treat telemetry as surveillance by default. Recommend layered visibility:
Flag any design that stores raw prompts, responses, secrets, personal data, or customer data. Require an explicit consent, retention, redaction, and access model before recommending raw-content logging.
Return:
npx claudepluginhub caravanai/caravan-tools --plugin ai-telemetryProvides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Fetches up-to-date documentation from Context7 for libraries and frameworks like React, Next.js, Prisma. Use for setup questions, API references, and code examples.