Deploy Now

SaaS Alternative

Helicone Portkey Langfuse Arize Phoenix

Categories

self-hosted developer-tools mcp-server ai-observability coding-agents token-tracking session-analytics

Stars

5,943

Forks

677

Watchers

33

Developer links

Last Updated: 09-20-2026
Share on X

AgentsView

Software developers running Claude Code, Cursor, Codex, and Aider use AgentsView to trace token expenditures, search full session transcripts, and analyze multi-agent concurrency timelines. Engineers can search through historical developer sessions to locate specific code modifications, prompt chains, or debugging rationales across extensive programming workflows. The real-time activity timeline visualizes concurrent agent operations, revealing peak execution windows, task run durations, and automated tool invocations. The usage analytics engine calculates financial costs and token volumes per model by applying live pricing tables while accounting for prompt caching read discounts. Developers inspect granular turn-by-turn session transcripts complete with raw model payloads, file difference diffs, executed shell commands, and syntax errors. The recall browser extracts key technical context from past programming sessions, organizing facts into an indexed reference knowledge corpus for reuse across future project tasks. Administrators can replicate SQLite session tables into DuckDB analytics files or PostgreSQL warehouses to query longitudinal metrics using SQL analytical tools or custom reporting scripts. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

AgentsView
AgentsView
AgentsView
AgentsView
AgentsView

Benefits

  • Unified Multi-Agent Transcript Full-Text Search
  • Discovers and indexes sessions from over twenty AI coding agents into a local SQLite repository, enabling instant cross-project keyword and semantic search across commands, file edits, and prompts.
  • Accurate Token Cost and Usage Accounting
  • Calculates precise financial expenditures using dynamic pricing tables, factoring in prompt cache generation and cache read discounts across daily, weekly, and per-model consumption windows.
  • Multi-Agent Concurrency and Activity Observability
  • Renders real-time Gantt timelines of overlapping agent workloads, allowing developers to monitor active runtimes, track tool invocation latency, and evaluate multi-agent orchestration efficiency.
  • Dual Engine DuckDB Analytical Mirroring
  • Mirrors session archives automatically into embedded DuckDB and external PostgreSQL databases, enabling developers to execute advanced analytical SQL queries against prompt histories and token distributions.

Features

  • Universal Agent Session Ingestion
  • Monitors local filesystem paths to ingest session transcripts from Claude Code, Codex, Cursor, Devin CLI, Goose, and Aider automatically.
  • Prompt Cache Aware Calculations
  • Distinguishes between input tokens, cache write tokens, cache read tokens, and output tokens to produce accurate cost estimates using LiteLLM rate cards.
  • Recall Knowledge Corpus Browser
  • Extracts architectural decisions, code explanations, and debugging learnings from session transcripts, cataloging them into a reusable knowledge base.
  • Turn-by-Turn Transcript Viewer
  • Displays complete conversational histories including model reasoning traces, executed bash commands, file diff modifications, and API payloads with syntax highlighting.
  • Model Context Protocol Integration
  • Exposes stored session archives and transcript search capabilities as a Model Context Protocol server for direct querying by external AI clients.
  • Embedded DuckDB Query Engine
  • Includes an embedded DuckDB analytical database for running high-speed columnar queries and exporting session metrics to Parquet and CSV formats.