Installation

PasteGuard ships as a single all-in-one Docker image — the Bun proxy and the PII detector (deterministic checks plus a semantic backend) run together in one container. GLiNER is included as the default backend. No build step required.

Docker Image

ghcr.io/sgasser/pasteguard

Tag	Description
`latest`	Latest release
`X.Y.Z`	Pinned release version

Detection is multilingual and language-agnostic — one model finds entities regardless of the language the text is written in, so there are no per-language image variants.

Quick Start

docker run --rm -p 3000:3000 ghcr.io/sgasser/pasteguard:latest

PasteGuard runs on http://localhost:3000. Open http://localhost:3000 for the dashboard.

The first start loads the detection model and can take a little longer; the container reports healthy once both the proxy and the detector are ready.

Detector startup crash in 0.9.3

The 0.9.3 image can terminate the detector with SIGSEGV while importing ONNX Runtime 1.29.0. Its telemetry initialization fails in the runtime image, which has no shell. The proxy then reports HTTP 503 from /health. For an existing 0.9.3 deployment, disable ONNX Runtime telemetry:

docker run --rm -p 3000:3000 -e ORT_DISABLE_TELEMETRY=1 \
  ghcr.io/sgasser/pasteguard:0.9.3

With Docker Compose, add this to the pasteguard service and recreate the container with docker compose up -d --force-recreate pasteguard:

environment:
  ORT_DISABLE_TELEMETRY: "1"

Images built from the current Dockerfile set this automatically. GLiNER detection remains enabled. The build also checks real detector startup and inference in the offline runtime as the same nonroot user that runs the service.

Production Setup

For persistent logs and custom configuration:

# Download config
curl -O https://raw.githubusercontent.com/sgasser/pasteguard/main/config.example.yaml
mv config.example.yaml config.yaml

# Create data directory
mkdir -p data

# Run with persistence
docker run -d --name pasteguard --restart unless-stopped -p 3000:3000 \
  -v ./config.yaml:/pasteguard/config.yaml:ro \
  -v ./data:/pasteguard/data \
  ghcr.io/sgasser/pasteguard:latest

This gives you:

Custom config — edit config.yaml to change detection settings
Persistent logs — request history in data/pasteguard.db survives restarts
Auto-restart — the container restarts automatically

Or with Docker Compose:

curl -O https://raw.githubusercontent.com/sgasser/pasteguard/main/docker-compose.yml
docker compose up -d

Languages and phone numbers

Detection is multilingual and needs no per-language configuration. Configure phone_regions only if you need national-format phone numbers without a + country prefix — see PII Detection Config.

Detector Models

The image includes urchade/gliner_multi_pii-v1 and loads it offline by default. Custom Hugging Face and local models are documented under GLiNER.

Environment Variables

Variable	Default	Description
`DETECTOR_URL`	`http://localhost:5002`	Where the proxy reaches the detector. In the all-in-one image this is in-container and rarely changed; override it to point at an external detector.
`DETECTOR_TIMEOUT`	`30`	Seconds to wait for each detector `/analyze` request when using the example config. Increase for very large messages; set to `0` to disable.
`DETECTOR_BACKEND`	`gliner`	Semantic detector backend. GLiNER is currently the only supported value.
`DETECTOR_MODEL`	`urchade/gliner_multi_pii-v1`	Hugging Face model ID or local model directory used by the detector.
`ORT_DISABLE_TELEMETRY`	`1` (Docker)	Disables ONNX Runtime telemetry before import; required by ONNX Runtime 1.29.0 in the shell-free runtime.
`PASTEGUARD_STARTUP_TIMEOUT`	`180`	Seconds to wait for the detector to become ready at startup

Getting Started

Use Cases

Concepts

API Reference

Configuration

Docker Image

Quick Start

Detector startup crash in 0.9.3

Production Setup

Languages and phone numbers

Detector Models

Environment Variables

Next Steps

Quickstart

Configuration

​Docker Image

​Quick Start

​Detector startup crash in 0.9.3

​Production Setup

​Languages and phone numbers

​Detector Models

​Environment Variables

​Next Steps

Quickstart

Configuration

Docker Image

Quick Start

Detector startup crash in 0.9.3

Production Setup

Languages and phone numbers

Detector Models

Environment Variables

Next Steps