vibelens 0.9.0

Agent Trajectory visualization and analysis platform

VibeLens

See what your AI coding agents are actually doing.

Live Demo · Quick Start · PyPI · Changelog

---

VibeLens parses, visualizes, and analyzes conversation histories from Claude Code, Codex CLI, Gemini CLI, and Dataclaw — giving you full observability into your AI-assisted development workflow.

One command. No config. Works with your local ~/.claude/ sessions out of the box.

pip install vibelens && vibelens serve

Analytics Dashboard Aggregate stats, cost estimation, and usage trends over time.	Activity Heatmap & Charts Peak hours, project breakdown, model and tool distribution.
Session Browser Browse by project, view step-by-step timelines with tool calls.	Conversation Flow Phase detection across exploration, implementation, and verification.

Why VibeLens?

• What did my agent actually do? Step-by-step timeline with tool calls, token counts, and elapsed time
• How much is it costing me? Per-session and aggregate cost estimation across 45+ models from 12 providers
• Where are the bottlenecks? Conversation flow diagrams with phase detection and tool dependency graphs
• How do I share a session? Shareable permalink URLs with read-only session replay
• What are my usage trends? Analytics dashboard with heatmaps, model distribution, and project breakdowns

Features

Feature	Description
Multi-agent parsing	Claude Code, Codex CLI, Gemini CLI, Dataclaw with auto-detection
Step timeline	Tool calls, sub-agent spawns, elapsed time, image content
Cost estimation	45+ models, 12 providers, per-session and aggregate cost
Flow diagrams	Phase-grouped conversation flow with tool dependency highlighting
Session sharing	Shareable permalink URLs with read-only view
Analytics dashboard	Stat cards, usage trends, activity heatmap, model distribution
Tool distribution	Per-tool call counts, error rates, avg/session

Quick Start

Install and run

pip install vibelens
vibelens serve

Or run without installing:

uvx vibelens serve

VibeLens opens your browser and reads your local ~/.claude/ sessions by default.

Development install

git clone https://github.com/yejh123/VibeLens.git
cd VibeLens
uv sync --extra dev
uv run vibelens serve

Configuration

YAML-based configuration with environment variable overrides (VIBELENS_*). See config/vibelens.example.yaml for all options.

# Use a config file
vibelens serve --config config/self-use.yaml

# Override host/port
vibelens serve --host 0.0.0.0 --port 8080

Supported Agents

Agent	Format	Data Location
Claude Code	JSONL	~/.claude/projects/
Codex CLI	JSONL	~/.codex/sessions/
Gemini CLI	JSON	~/.gemini/tmp/
Dataclaw	JSONL	HuggingFace exports

Data Donation

VibeLens supports donating your agent conversation data to advance research on coding agent behavior. Donated sessions are collected by CHATS-Lab (Conversation, Human-AI Technology, and Safety Lab) at Northeastern University.

To donate, upload your data, select the sessions you want to share, and click the Donate button.

Development

# Lint and test
uv run ruff check src/ tests/
uv run pytest tests/ -v -s

# Frontend dev server (hot reload)
cd frontend && npm install && npm run dev

Contributing

Contributions are welcome! Please ensure code passes ruff check and pytest before submitting.

License

MIT