
Tma1
- 114 repo stars
- Updated August 1, 2026
- tma1-ai/tma1
Local-first LLM agent observability. Track tokens, cost, and conversations on your machine.
About
tma1 is a Claude Code skill in the AI & Agent Building category. Local-first LLM agent observability. Track tokens, cost, and conversations on your machine.
- tma1
- AI & Agent Building
- AI-coding skill
Tma1 by the numbers
- Data as of Aug 2, 2026 (Skillselion catalog sync)
/plugin marketplace add tma1-ai/tma1/plugin install tma1@tma1Add your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| repo stars | ★ 114 |
|---|---|
| Last updated | August 1, 2026 |
| Repository | tma1-ai/tma1 ↗ |
What it does
Local-first LLM agent observability. Track tokens, cost, and conversations on your machine.
README.md
TMA1 Claude Code Plugin
Local-first LLM agent observability for Claude Code.
Installation
From Claude Code Marketplace (recommended)
/install-plugin tma1
Manual
Clone this repository and add the plugin path to your Claude Code configuration:
git clone https://github.com/tma1-ai/tma1.git
Add to ~/.claude/settings.json:
{
"plugins": ["path/to/tma1/claude-plugin"]
}
Skills
/tma1-setup — Install and configure TMA1
Installs the tma1-server binary, starts it, and configures your OTel exporter endpoint.
/tma1-setup
/tma1 — Query observability data
Ask questions about your agent's behavior without opening the dashboard.
/tma1 how much did I spend today?
/tma1 show me recent traces
/tma1 any errors in the last hour?
/tma1 compare model costs this week
How it works
TMA1 captures OpenTelemetry metrics and logs emitted by Claude Code (and traces from other agents) and stores them in a local GreptimeDB instance. The plugin provides two skills:
- tma1-setup handles the one-time installation and configuration
- tma1 queries the collected data via SQL and presents results inline
All data stays on your machine. No cloud, no accounts, no API keys.
OTel Configuration
Claude Code exports metrics and logs (not traces). Add to ~/.claude/settings.json:
{
"env": {
"OTEL_EXPORTER_OTLP_ENDPOINT": "http://localhost:14318/v1/otlp",
"OTEL_EXPORTER_OTLP_PROTOCOL": "http/protobuf",
"OTEL_METRICS_EXPORTER": "otlp",
"OTEL_LOGS_EXPORTER": "otlp"
}
}
Requirements
- Claude Code with OTel support enabled
- macOS (arm64/amd64) or Linux (amd64/arm64)
- ~200 MB disk space for GreptimeDB