Metadata-Version: 2.4
Name: tacit-ai
Version: 0.1.1rc4
Summary: CLI for evidence-grounded incident investigation across observability tools.
License: MIT
License-File: LICENSE
Requires-Python: >=3.12
Requires-Dist: aiohttp>=3.14.1
Requires-Dist: anthropic>=0.49.0
Requires-Dist: click>=8.1.0
Requires-Dist: fastapi>=0.111.0
Requires-Dist: httpx>=0.27.0
Requires-Dist: openai>=1.35.0
Requires-Dist: promql-parser>=0.9.0
Requires-Dist: pydantic-settings>=2.14.2
Requires-Dist: pydantic>=2.7.0
Requires-Dist: python-dotenv>=1.0.1
Requires-Dist: pyyaml>=6.0
Requires-Dist: rich>=13.0.0
Requires-Dist: slack-bolt>=1.19.0
Requires-Dist: slack-sdk>=3.31.0
Requires-Dist: structlog>=24.2.0
Requires-Dist: tenacity>=8.4.0
Requires-Dist: truststore>=0.10.4
Requires-Dist: uvicorn[standard]>=0.30.0
Provides-Extra: bedrock
Requires-Dist: boto3>=1.34.116; extra == 'bedrock'
Description-Content-Type: text/markdown

# Tacit CLI

Tacit helps operators turn incident prompts and existing operational context into evidence-grounded investigations from the command line.

The `tacit` command can:

- configure local Grafana, SignalFx, and LLM settings
- run environment health checks
- run an investigation workflow from a natural-language prompt
- start the local API and browser UI
- run the demo stack
- ingest existing dashboards, alerts, runbooks, and incidents as reusable operational context
- export an anonymized assessment bundle for sharing

## Install

```bash
uvx --from tacit-ai tacit --help
```

Or install into your environment:

```bash
pip install tacit-ai
tacit --help
```

## Quick Start

Configure Tacit:

```bash
tacit setup
```

Check your local environment:

```bash
tacit doctor
```

Run a test investigation:

```bash
tacit test --prompt "checkout-service p95 latency is high"
```

Run the local API and UI:

```bash
tacit serve --host 127.0.0.1 --port 8000 --no-slack
```

Then open `http://127.0.0.1:8000`.

## Common Commands

```bash
tacit setup
tacit doctor
tacit test --prompt "5xx errors on checkout-service"
tacit serve --no-slack
tacit demo
tacit learn dashboard <dashboard_uid>
tacit learn approve <dashboard_uid>
tacit history list
tacit export-report --anonymous --validate
```

## Configuration

Tacit reads settings from environment variables, `.env`, and optional YAML config.
For local development, start from the repository's `.env.example` or `tacit.yaml.example`.

Useful settings include:

- `GRAFANA_URL`
- `GRAFANA_API_KEY`
- `LLM_PROVIDER`
- `LLM_MODEL`
- `LLM_API_KEY`
- `SIGNALFX_API_TOKEN`
- `API_AUTH_ENABLED`
- `API_AUTH_KEY`

## More Documentation

The full repository README covers architecture, development workflows, Docker compose stacks, evaluation notes, and project roadmap:

https://github.com/aditki/tacit
