Track usage, costs, and efficiency across AI coding agents
Project description
Track usage, costs, and efficiency across AI coding agents.
Supported Agents
| Agent | Log Location | Status |
|---|---|---|
| Claude Code | ~/.claude/projects/ |
Supported |
| Gemini CLI | ~/.gemini/tmp/<hash>/chats/ |
Supported |
| OpenAI Codex | ~/.codex/sessions/ |
Supported |
aidash auto-detects which agents are installed. No API keys, no config.
Installation
Python engine (required):
pip install aidash
Ink terminal UI (optional, for interactive views):
npm install -g aidash-ui
Development
git clone https://github.com/kaustubhraiiii/aidash.git
cd aidash
pip install -e .
cd frontend && npm install && npm run build
Quick Start
CLI (Python engine):
aidash # show available commands
aidash cost --period weekly # last 7 days of spend
aidash score last # rate your most recent session
aidash search "auth" # full-text search across sessions
Ink interactive views (requires aidash-ui):
aidash-ui cost # interactive cost dashboard
aidash-ui replay last # scrollable session timeline
aidash-ui score --trend # efficiency score with trend
aidash-ui rates --compare # model pricing comparison
aidash-ui search "auth" # live search results
JSON output (pipe-friendly, same data as views):
aidash-ui cost --json # structured JSON envelope
aidash-ui search "auth" --json # use in scripts
Commands
cost
Unified cost dashboard across all agents.
aidash cost
aidash cost --period monthly --by agent
aidash cost --agent claude_code
aidash cost --export csv # export to file
replay
Play back a coding session as a scrollable terminal timeline.
aidash replay last
aidash replay today
aidash replay a1b2c3d4
score
Rate your sessions on a 0-100 efficiency scale.
aidash score last
aidash score --trend
aidash score today
rates
Compare pricing across models and agents.
aidash rates
aidash rates --compare
aidash rates --period monthly
search
Full-text search across all sessions.
aidash search "auth migration"
aidash search --agent claude_code "database"
aidash search --project myapp "OAuth"
Claude Code Plugin
Install the aidash slash commands directly into Claude Code:
claude /plugin marketplace add kaustubhraiiii/aidash-cli
This adds five slash commands to your Claude Code sessions:
| Command | Description |
|---|---|
/aidash:cost |
Cost breakdown with optional period/agent/project filters |
/aidash:replay |
Session timeline playback |
/aidash:score |
Efficiency score with trend analysis |
/aidash:rates |
Per-model pricing from real usage |
/aidash:search |
Full-text session search |
Requires pip install aidash with Python 3 on PATH.
Scoring Methodology
| Metric | Weight | What it measures |
|---|---|---|
| Prompt-to-completion ratio | 30% | Clarity of instructions — fewer prompts per agent reply. |
| Tool call efficiency | 25% | Diverse, purposeful tool use vs. repetitive thrashing. |
| Token density | 25% | Average meaningful output per agent turn. |
| Session focus | 20% | Sweet spot of 3–8 distinct tools per session. |
How It Works
- Detection — Scans for
~/.claude/,~/.gemini/tmp/, and~/.codex/. - Parsing — Reads JSONL/JSON session files for messages, tokens, tool calls, timestamps.
- Normalization — Maps each agent's format into a common model (Session, Message, TokenUsage, ToolCall).
- Analysis — Computes costs, efficiency scores, and what-if comparisons. Strictly read-only.
Tech Stack
- Python 3.10+ — data engine, session parsing, cost calculation
- Click 8.0+ — CLI framework
- TypeScript / Node 18+ — Ink frontend
- Ink — React-based terminal UI
- termcn — terminal component library
Contributing
- Fork the repository
- Create a feature branch (
git checkout -b feat/my-feature) - Commit your changes (
git commit -am 'Add my feature') - Push the branch (
git push origin feat/my-feature) - Open a Pull Request
Bugs and feature requests welcome via Issues.
License
MIT © 2026 Kaustubh Rai
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file aidash-0.2.1.tar.gz.
File metadata
- Download URL: aidash-0.2.1.tar.gz
- Upload date:
- Size: 33.1 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.14.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
bfaa5aa2d6aa1483a27c22da47f4573e4f13f0b449bd2460ec99d8333bff7f1a
|
|
| MD5 |
50384e6d4f09d8f30ad1220554f393e9
|
|
| BLAKE2b-256 |
2f271395ecfa9518aab2ad4d293102ccc24a21f8fba0618b5b4009f6b4f13228
|
File details
Details for the file aidash-0.2.1-py3-none-any.whl.
File metadata
- Download URL: aidash-0.2.1-py3-none-any.whl
- Upload date:
- Size: 35.4 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.14.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
e328a457d32858f9f96236becf4423e2b3160ce2bd3116e2cffe678dbbe106f2
|
|
| MD5 |
45e2aec6fa3ea25e9af3c9056d54501a
|
|
| BLAKE2b-256 |
33fd88e26d154ea35fb4a6e086e70d861d3c0c2b311d8dd18a6845511eddfadf
|