docs(readme): refactor with comparative benchmark + logo

Add Dockermint logo, new badge row (License, Website, Docs, Go MSRV,
Release, CI, Release workflow, Container), 3-way comparative benchmark
table (LevelDB native / Level2Pebble / Pebblify), restructured install
section (Docker/Podman recommended, Binary, Build from source), beta
warning on daemon mode, improved intro and CTA. Removed Artifact
attestation section and early-stage warning.

Closes #55

Author: nayrosk

README.md

@@ -1,92 +1,142 @@
+<p align="center">
+  <img src="./assets/logo.svg" alt="Pebblify" width="200" />
+</p>
+
 # Pebblify
 
 [![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
-[![Go](https://img.shields.io/badge/Go-1.25+-00ADD8?logo=go&logoColor=white)](https://go.dev/)
-![Benchmark](https://img.shields.io/badge/benchmark-217M_keys_in_4m11s-brightgreen)
-![Throughput](https://img.shields.io/badge/throughput-865k_keys%2Fs-blue)
-
-**Pebblify** is a high-performance migration tool that converts LevelDB databases to PebbleDB format, specifically designed for Cosmos SDK and CometBFT (formerly Tendermint) blockchain nodes.
-
-PebbleDB offers significant performance improvements over LevelDB, including better write throughput, more efficient compaction, and reduced storage overhead. Pebblify makes it easy to migrate your existing node data without manual intervention.
+[![Website](https://img.shields.io/badge/website-dockermint.io-informational)](https://dockermint.io/tools)
+[![Documentation](https://img.shields.io/badge/docs-docs.dockermint.io-blue)](https://docs.dockermint.io/pebblify/)
+[![Go MSRV](https://img.shields.io/badge/Go-1.25+-00ADD8?logo=go&logoColor=white)](https://go.dev/doc/install)
+[![Latest Release](https://img.shields.io/github/v/release/Dockermint/pebblify?label=release)](https://github.com/Dockermint/pebblify/releases)
+[![CI](https://github.com/Dockermint/pebblify/actions/workflows/ci.yml/badge.svg)](https://github.com/Dockermint/pebblify/actions/workflows/ci.yml)
+[![Release](https://github.com/Dockermint/pebblify/actions/workflows/release.yml/badge.svg)](https://github.com/Dockermint/pebblify/actions/workflows/release.yml)
+[![Container](https://img.shields.io/badge/container-ghcr.io-blue?logo=github)](https://github.com/orgs/Dockermint/packages/container/package/pebblify)
 
-📖 [Documentation](https://docs.dockermint.io/pebblify/) · 🌐 [Website](https://dockermint.io/tools)
+**Pebblify** is a high-performance migration tool that converts LevelDB databases to PebbleDB format, specifically designed for Cosmos SDK and CometBFT blockchain nodes.
 
-> [!WARNING]
-> This tool is still in the early stages of development and may contain bugs or be unstable. If you notice any unusual behavior, please open an issue.
+PebbleDB offers significant performance improvements over LevelDB, including better write throughput, more efficient compaction, and reduced storage overhead. Pebblify outperforms existing tools: in a real-world production benchmark it completed a full snapshot conversion in **6m48** versus 7m13 for Level2Pebble, while producing a **20.34 GiB** output compared to 26 GiB, with no data loss. It adds crash recovery, disk space checks, verification, Prometheus metrics, health probes, notifications, and daemon mode that Level2Pebble does not provide.
 
-## Features
+**[Install now](#installation)** or read the [full documentation](https://docs.dockermint.io/pebblify/).
 
-- **Fast parallel conversion** — Process multiple databases concurrently with configurable worker count
-- **Crash recovery** — Resume interrupted migrations from the last checkpoint
-- **Adaptive batching** — Automatically adjusts batch sizes based on memory constraints
-- **Real-time progress** — Live progress bar with throughput metrics and ETA
-- **Data verification** — Verify converted data integrity with configurable sampling
-- **Disk space checks** — Pre-flight validation to ensure sufficient storage
-- **Docker support** — Multi-architecture container images (amd64/arm64)
-- **Health probes** — HTTP liveness, readiness, and startup endpoints for orchestrators
-- **Prometheus metrics** — Opt-in metrics exporter for conversion monitoring
-- **Shell completion** — Bash and zsh autocompletion via `pebblify completion`
+## Quick Start
 
-## Requirements
+### Convert a LevelDB database
 
-- **Go 1.25+** (for building from source)
-- **Sufficient disk space** — Approximately 1.5x the source data size during conversion
-- **Source database** — Valid LevelDB directory structure (Cosmos/CometBFT `data/` format)
+```bash
+pebblify level-to-pebble ~/.gaia/data ./gaia-pebble
+```
 
-## Installation
+### Recover an interrupted conversion
 
-### CLI (all platforms)
+```bash
+pebblify recover --tmp-dir /var/tmp
+```
 
-Cross-platform. Works on `linux/amd64`, `linux/arm64`, `darwin/amd64`, and `darwin/arm64`.
+### Verify converted data
 
 ```bash
-git clone https://github.com/Dockermint/pebblify.git
-cd pebblify
-make install-cli   # build and install pebblify binary to PATH
+pebblify verify --sample 10 ~/.gaia/data ./gaia-pebble/data
 ```
 
-Or download a pre-built binary from [releases](https://github.com/Dockermint/pebblify/releases).
+For all available flags and subcommands, see the [CLI reference](https://docs.dockermint.io/pebblify/).
 
-### Systemd daemon — **Linux only**
+## Benchmark
 
-Installs the daemon as a system service. Requires root. Not supported on macOS.
+Real-world comparison on a production Cosmos node snapshot:
+
+| Metric | LevelDB (native) | [Level2Pebble](https://github.com/notional-labs/level2pebble) | Pebblify |
+|---|---|---|---|
+| Size | 25 GiB | 26 GiB | **20.34 GiB** |
+| Duration | -- | 7m13 | **6m48** |
+| DB performance | average | good | good |
+| Full snapshot conversion | -- | ❌ | ✅ |
+| Disk space checks | -- | ❌ | ✅ |
+| Prometheus metrics | -- | ❌ | ✅ |
+| Probes | -- | ❌ | ✅ |
+| Resume on error | -- | ❌ | ✅ |
+| Verification | -- | ❌ | ✅ |
+| Daemon mode | -- | ❌ | ✅ |
+| Notifications* | -- | ❌ | ✅ |
+| Push* | -- | ❌ | ✅ |
+
+\* Daemon mode only.
 
-```bash
-sudo make install-systemd-daemon
-```
+## Features
 
-After installation, fill in `/etc/pebblify/.env`, then:
+- **Fast parallel conversion** -- Process multiple databases concurrently with configurable worker count
+- **Crash recovery** -- Resume interrupted migrations from the last checkpoint
+- **Adaptive batching** -- Automatically adjusts batch sizes based on memory constraints
+- **Real-time progress** -- Live progress bar with throughput metrics and ETA
+- **Data verification** -- Verify converted data integrity with configurable sampling
+- **Disk space checks** -- Pre-flight validation to ensure sufficient storage
+- **Docker support** -- Multi-architecture container images (amd64/arm64)
+- **Health probes** -- HTTP liveness, readiness, and startup endpoints for orchestrators
+- **Prometheus metrics** -- Opt-in metrics exporter for conversion monitoring
+- **Shell completion** -- Bash and zsh autocompletion via `pebblify completion`
 
-```bash
-sudo systemctl daemon-reload
-sudo systemctl enable --now pebblify
-```
+## Installation
 
-See [daemon quickstart](docs/markdown/daemon-quickstart.md) for the full setup guide.
+### Docker / Podman (recommended)
 
-### Podman Quadlet (rootless)
+> [!TIP]
+> Podman is rootless, giving it a security advantage over Docker.
 
-Deploys the daemon as a rootless Podman container managed by the systemd user session. Linux-native; macOS users need Podman Desktop.
+**Prerequisites:** [Docker](https://docs.docker.com/get-docker/) or [Podman](https://podman.io/docs/installation).
 
 ```bash
-make install-podman
-```
+# Pull the image
+docker pull ghcr.io/dockermint/pebblify:0.4.0
+podman pull ghcr.io/dockermint/pebblify:0.4.0
 
-After installation:
+# Run a conversion
+docker run --rm \
+  -v /path/to/source:/data/source:ro \
+  -v /path/to/output:/data/output \
+  -v /path/to/tmp:/tmp \
+  ghcr.io/dockermint/pebblify:0.4.0 \
+  level-to-pebble --health --metrics /data/source /data/output
 
-```bash
-systemctl --user daemon-reload
-systemctl --user start pebblify
+podman run --rm \
+  -v /path/to/source:/data/source:ro \
+  -v /path/to/output:/data/output \
+  -v /path/to/tmp:/tmp \
+  ghcr.io/dockermint/pebblify:0.4.0 \
+  level-to-pebble --health --metrics /data/source /data/output
 ```
 
-### Docker
+### Binary
+
+Download a pre-built binary for your platform from the [v0.4.0 release page](https://github.com/Dockermint/pebblify/releases/tag/v0.4.0).
+
+Available architectures: Linux/AMD64, Linux/ARM64, Darwin/ARM64, Darwin/AMD64.
+
+After downloading, verify the checksum against the `checksums.txt` file on the release page, then place the binary on your `PATH`.
+
+### Build from source
+
+> [!NOTE]
+> **Prerequisites:** [Go 1.25+](https://go.dev/doc/install), `make`, `git`.
+>
+> Per-platform Makefile targets:
+> - `install-cli` -- build and install the CLI binary to your `PATH` (all platforms)
+> - `build-docker` -- build a Docker image for the local platform
+> - `install-systemd-daemon` -- install the daemon as a systemd service (Linux only, requires root)
+> - `install-podman` -- install the daemon as a rootless Podman Quadlet (Linux only)
 
 ```bash
-make build-docker
+git clone --branch v0.4.0 https://github.com/Dockermint/pebblify.git
+cd pebblify
+make install-cli
 ```
 
+For daemon and container installation options, see the [installation guide](https://docs.dockermint.io/pebblify/).
+
 ## Daemon mode
 
+> [!WARNING]
+> The daemon mode is currently in **beta**. If you encounter a bug, please [open an issue](https://github.com/Dockermint/pebblify/issues/new/choose).
+
 `pebblify daemon` is a long-running HTTP service that accepts snapshot archive URLs, converts them from LevelDB to PebbleDB format, repacks the output, and saves it to one or more storage targets (local directory, SCP, S3). Jobs are submitted via a REST API and processed serially.
 
 **Platform:** The `daemon` subcommand is Linux-only at runtime. On macOS, use Docker Compose or Podman.
@@ -109,92 +159,60 @@ curl -s -X POST http://127.0.0.1:2324/v1/jobs \
 
 Full guide: [docs/markdown/daemon-quickstart.md](docs/markdown/daemon-quickstart.md)
 
-## Quick Start
+## Systemd daemon (Linux only)
 
-### Convert
+Installs the daemon as a system service. Requires root. Not supported on macOS.
 
 ```bash
-pebblify level-to-pebble ~/.gaia/data ./gaia-pebble
+sudo make install-systemd-daemon
 ```
 
-### Recover an Interrupted Conversion
+After installation, fill in `/etc/pebblify/.env`, then:
 
 ```bash
-pebblify recover --tmp-dir /var/tmp
+sudo systemctl daemon-reload
+sudo systemctl enable --now pebblify
 ```
 
-### Verify
+See [daemon quickstart](docs/markdown/daemon-quickstart.md) for the full setup guide.
 
-```bash
-pebblify verify --sample 10 ~/.gaia/data ./gaia-pebble/data
-```
+## Podman Quadlet (rootless)
 
-### Docker
+Deploys the daemon as a rootless Podman container managed by the systemd user session. Linux-native; macOS users need Podman Desktop.
 
 ```bash
-docker pull ghcr.io/dockermint/pebblify:latest
-
-docker run --rm \
-  -v /path/to/source:/data/source:ro \
-  -v /path/to/output:/data/output \
-  -v /path/to/tmp:/tmp \
-  ghcr.io/dockermint/pebblify:latest \
-  level-to-pebble --health --metrics /data/source /data/output
+make install-podman
 ```
 
-> For full command reference and all available flags, see the [documentation](https://docs.dockermint.io/pebblify/).
-
-## Artifact attestation
-
-Starting with v0.4.0, every release binary and Docker image is published with SLSA provenance and SBOM attestations. Verify before running:
+After installation:
 
 ```bash
-# Verify a release binary
-gh attestation verify pebblify-linux-amd64 \
-  --repo Dockermint/Pebblify
-
-# Verify the Docker image
-gh attestation verify \
-  oci://ghcr.io/dockermint/pebblify:v0.4.0 \
-  --repo Dockermint/Pebblify
+systemctl --user daemon-reload
+systemctl --user start pebblify
 ```
 
-Available attestation targets: `pebblify-linux-amd64`, `pebblify-linux-arm64`, `pebblify-darwin-amd64`, `pebblify-darwin-arm64`, and the multi-arch Docker image.
-
-Full guide: [docs/markdown/release-automation.md](docs/markdown/release-automation.md)
-
-## Benchmark
-
-Real-world conversion on a production Cosmos node dataset:
-
-| Metric | Value |
-|---|---|
-| Total keys | 217,895,735 |
-| Duration | 4m 11s |
-| Throughput | ~865k keys/s · ~160 MB/s |
-| Data processed | 39.54 GiB read / 39.54 GiB written |
-| Size reduction | -16.3% (LevelDB 23.32 GiB → PebbleDB 19.51 GiB) |
-| Data loss | None — 1:1 write/read parity |
+## Performance Tips
 
-> [!NOTE]
-> Benchmark performed on AMD Ryzen 9 8940HX, 32 GiB DDR5, NVMe (Btrfs). Temp folder on disk, not in RAM.
+- **Use SSDs** -- NVMe storage significantly improves conversion speed
+- **Increase workers** -- For systems with many CPU cores, increase `-w` for faster parallel processing
+- **Adjust batch memory** -- Increase `--batch-memory` if you have RAM to spare
+- **Use local temp** -- If `/tmp` is a tmpfs (RAM-based), use `--tmp-dir` to point to disk storage for large datasets
 
-## Performance Tips
+## Requirements
 
-- **Use SSDs** — NVMe storage significantly improves conversion speed
-- **Increase workers** — For systems with many CPU cores, increase `-w` for faster parallel processing
-- **Adjust batch memory** — Increase `--batch-memory` if you have RAM to spare
-- **Use local temp** — If `/tmp` is a tmpfs (RAM-based), use `--tmp-dir` to point to disk storage for large datasets
+- **Go 1.25+** (for building from source)
+- **Sufficient disk space** -- Approximately 1.5x the source data size during conversion
+- **Source database** -- Valid LevelDB directory structure (Cosmos/CometBFT `data/` format)
 
 ## Contributing
 
-Contributions are welcome! Please feel free to submit issues and pull requests.
+Contributions are welcome. Please submit issues and pull requests via [GitHub](https://github.com/Dockermint/pebblify).
 
 ## License
 
-This project is licensed under the Apache License 2.0 - see the [LICENSE](LICENCE) file for details.
+This project is licensed under the Apache License 2.0. See the [LICENSE](LICENCE) file for details.
 
 ## Acknowledgments
 
-- [CockroachDB Pebble](https://github.com/cockroachdb/pebble) — The high-performance storage engine
-- [syndtr/goleveldb](https://github.com/syndtr/goleveldb) — LevelDB implementation in Go
+- [CockroachDB Pebble](https://github.com/cockroachdb/pebble) -- The high-performance storage engine
+- [syndtr/goleveldb](https://github.com/syndtr/goleveldb) -- LevelDB implementation in Go