Every premium and members-only resource for $9.99/month, as a no-commitment subscription. See the plans

PalateKit

Sign in
Menu

Request a resource

Describe the resource you are missing in one sentence. We will email you when it is published.

0 / 280

A page, a video or an image online.

Unlock PalateKit

Unlock everything

Plan

Paying means accepting our terms of sale.

Secure payment by Stripe. Receipts and invoices sent by email.

Welcome back

Sign in to install, download or save this resource. A free account is enough for free resources.

Tracking AI model calls with PostHog v1.0.0

Use this resource to trace AI model usage with token counts, latency and costs

New

  • 379 downloads
  • 181 likes
  • 181 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install llm-usage-analytics-instrumentation

With no option, the skill goes to ~/.agents/skills/; add --agent claude-code (or codex, cursor…) to target an agent.

First time: connect the CLI with

npx --yes palatekit@latest login

The agent finds the calls to model providers in your code and adds generation capture with token counts, latency and cost. If the SDK is not installed, it also handles the initial setup.

Mostly picked by

Updated November 2, 2026

Card for the “Tracking AI model calls with PostHog” skill

Details

Works in
Install guides
Generates with
Your agent's model
Licence
MIT

Similar resources

View all

What you get

  • SKILL.md: steps for finding and capturing generations
  • references/: guides per provider (Anthropic, Google, Azure OpenAI, Cohere, DeepSeek, etc.) and cost calculation

Good fit for

  • You added a feature that calls an AI model and want to track its usage
  • You want to compare costs across providers or agent frameworks

Not for

    Tags

    • llm
    • tokens
    • costs
    • observability

    About

    You say

    “Trace the Anthropic calls of our writing assistant, with tokens and cost per request.”

    Card for the “Tracking AI model calls with PostHog” skill

    Skill published by posthog under the MIT license, delivered in its original version.

    Show more

    How to use it

    1. Install the skill in your agent's skills folder
    2. Ask your agent to trace AI model usage with token counts, latency and costs
    3. Review and adjust the result

    Content preview

    Files

    • LICENSE1.1 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • references/anthropic.md6.8 KB
    • references/autogen.md4.6 KB
    • references/azure-openai.md6.5 KB
    • references/basics.md3.3 KB
    • references/calculating-costs.md8.5 KB
    • references/cerebras.md6.9 KB
    • references/cohere.md6.9 KB
    • references/crewai.md4.6 KB
    • references/deepseek.md6.9 KB
    • references/dspy.md4.5 KB
    • references/fireworks-ai.md7 KB
    • references/google.md7.9 KB
    • references/groq.md6.9 KB
    • references/helicone.md7.1 KB
    • references/hugging-face.md7 KB
    • references/instructor.md6 KB
    • references/langchain.md7.1 KB
    • references/langgraph.md6.8 KB
    • references/litellm.md6.8 KB
    • references/llamaindex.md4.8 KB
    • references/manual-capture.md16 KB
    • references/mastra.md5.3 KB
    • references/mirascope.md4.6 KB
    • references/mistral.md6.9 KB
    • references/ollama.md6.9 KB
    • references/openai-agents.md5.9 KB
    • references/openai.md6.3 KB
    • references/openrouter.md7.3 KB
    • references/perplexity.md6.9 KB
    • references/portkey.md7.2 KB
    • references/pydantic-ai.md4.7 KB
    • references/semantic-kernel.md4.6 KB
    • references/smolagents.md4.4 KB
    • references/together-ai.md6.9 KB
    • references/traces.md5.9 KB
    • references/vercel-ai.md4.6 KB
    • references/xai.md6.8 KB
    • SKILL.md6.7 KB

    Origin and licence

    Listed unmodified

    Author
    posthog
    Source
    github.com/posthog/skills/tree/221bfed4f757a4b7e1826a9573665e1152c4d728/skills/posthog/llm-analytics/skills/all
    Licence
    MIT

    Notice

    « omnibus-instrument-llm-analytics » (posthog, MIT) livré dans sa version d'origine : fiche en français et en anglais par PalateKit.

    User reviews

    –

    No ratings yet

    Loading reviews…

    Delete your review?

    Your rating and comment will be removed from the page.