Agents Metrics

Tracks token usage, cost and sessions for LLM agent runs

Community 0 6/wk MIT v1.5.0

Installation

pnpm add @zetesis/payload-agents-metrics

About

Agents Metrics records every completed LLM agent run in Payload and works out what it cost, based on each model's token prices. Each run is stored with its user, agent, model, provider and conversation, so you can see spending and usage over time without building the bookkeeping yourself.

The admin gets a read-only log of usage events, and the plugin adds an endpoint set for totals, daily series, top spenders by cost and conversation lists with transcripts. A ready-made dashboard component shows KPI cards, a per-day chart and a sessions panel, and it can be filtered by dimension. In multi-tenant setups, reads are limited to the tenants the current user can access.

Package info

Package name
@zetesis/payload-agents-metrics
Latest version
1.5.0
Unpacked size
239 kB
License
MIT
Weekly downloads
6
Last publish
Jul 6, 2026

More from Zetesis-Labs

View profile

MCP server that gives AI tools search and retrieval over Typesense-indexed Payload content

Adds AI agent management, streaming chat endpoints and gateway-backed model access to Payload

Vector search adapter that stores embeddings in Postgres with pgvector

Keeps Payload collections synced to Typesense with keyword, semantic and hybrid search

Hierarchical taxonomy collection and relationship field for Payload collections

Builds Lexical rich text editors from feature and block lists, with typed block renderers

Similar plugins

More in AI

Visual drag-and-drop page builder for Payload collections with live canvas editing

Chat agent that composes and edits Payload pages from your block library

Translates default-locale content into secondary locales with OpenAI or a custom provider

Translates localized content fields with an AI provider from the admin panel

Works out EU VAT on Payload orders, with reverse charge and OSS totals

Writes product titles, descriptions and translations with Claude, ChatGPT, Gemini or Kimi