View Creator

View Creator (Viewkit) is a CLI tool that helps you initialize, manage, and publish Shinzo views.

1. Prerequisites

macOS on Apple Silicon (M1/M2/M3) or Linux (x86_64 / amd_64)

Tools

git
make
Go 1.23+ installed and on your PATH

Check Go:

go version
# expect something like: go version go1.23.x ...

Optional but helpful:

A code editor (VS Code, GoLand, etc.)
Basic comfortability running terminal commands

2. Clone the repository

From the directory where you keep your projects:

git clone https://github.com/shinzonetwork/shinzo-view-creator.git
cd shinzo-view-creator

3. Build the `viewkit` binary

From the repo root:

make build

If the build is successful, you should see a build directory:

ls build
# viewkit  (plus any other build artifacts)

You can now run Viewkit via:

./build/viewkit --help

Optional: add `viewkit` to your PATH

So you don’t have to type ./build each time:

echo 'export PATH="$PWD/build:$PATH"' >> ~/.zshrc
source ~/.zshrc

Then you can run:

viewkit --help

4. Wasmer runtime

Viewkit can execute WebAssembly lenses locally to validate and preview them.

Under the hood, it uses wasmer-go, which depends on a native dynamic library (libwasmer.dylib). If macOS cannot find that library, any command that touches lenses will fail with an error like:

image not found
library not loaded: libwasmer.dylib

This section explains how to set that up.

4.1 Install the Wasmer Go module

From the repo root:

go get github.com/wasmerio/wasmer-go@v1.0.4

This ensures wasmer-go and its packaged native libraries are present in your GOPATH.

4.2 Environment variables

We will use three environment variables:

WASMER_ROOT
Points to the directory where libwasmer.dylib lives.
WASMER_LIB_PATH
Used by wasmer-go to find the dynamic library.
DYLD_LIBRARY_PATH
macOS dynamic loader search path. We prepend WASMER_ROOT so the loader can find libwasmer.dylib when viewkit starts.

4.3 Configure the env vars

Append these lines to your ~/.zshrc if on Apple Silicon:

echo 'export WASMER_ROOT="$(go env GOPATH)/pkg/mod/github.com/wasmerio/wasmer-go@v1.0.4/wasmer/packaged/lib/darwin-aarch64"' >> ~/.zshrc
echo 'export WASMER_LIB_PATH="$WASMER_ROOT"' >> ~/.zshrc
echo 'export DYLD_LIBRARY_PATH="$WASMER_ROOT:$DYLD_LIBRARY_PATH"' >> ~/.zshrc

Otherwise append these lines to your ~/.zshrc if on Linux:

echo 'export WASMER_ROOT="$(go env GOPATH)/pkg/mod/github.com/wasmerio/wasmer-go@v1.0.4/wasmer/packaged/lib/linux-amd64"' >> ~/.zshrc
echo 'export WASMER_LIB_PATH="$WASMER_ROOT"' >> ~/.zshrc
echo 'export LD_LIBRARY_PATH="$WASMER_ROOT:$LD_LIBRARY_PATH"' >> ~/.zshrc

Reload your shell configuration:

source ~/.zshrc

Verify:

echo "$WASMER_ROOT"
ls "$WASMER_ROOT"
# you should see libwasmer.dylib here

If libwasmer.dylib is missing, re-run the go get step and ensure go env GOPATH returns a valid path.

5. Concepts: Views, queries, SDL, lenses, wallets

Viewkit revolves around views. Each view is a bundle that includes:

Metadata – name, version, target info
Query – the raw shape of data to ingest
(for example: Log { address topics data transactionHash blockNumber })
SDL (GraphQL) – how that data is modeled/exposed
(for example: type FilteredAndDecodedLogs { transactionHash: String })
Lenses (WASM) – chained transforms that filter/decode/reshape data
Wallet – key used to sign deployments to a target network (local, devnet, etc.)

Think of the pipeline as:

raw data → query → lenses (WASM) → GraphQL SDL → view stored + queryable

6. Quickstart: create and deploy a view (`testdeploy`)

We’ll walk end-to-end through an example view named testdeploy:

Initialize the view bundle
Inspect it
Add a query
Add SDL
Attach a lens
Generate a wallet
Deploy locally (with DefraDB Playground)
Deploy to devnet

All commands assume viewkit is on your PATH.
If not, replace viewkit with ./build/viewkit.

6.1 Initialize the view bundle

viewkit view init testdeploy

This:

Creates a new view bundle called testdeploy on disk
Registers internal metadata for queries, SDL, lenses, and versions

6.2 Inspect the bundle

viewkit view inspect testdeploy

You should see something like:

View name: testdeploy
A version / ID
Initially empty or default sections for:
- Queries
- SDL
- Lenses

6.3 Add a query (raw ingest shape)

Define the raw data shape to ingest, e.g. basic EVM logs:

viewkit view add query \
  "Log {address topics data transactionHash blockNumber}" \
  --name testdeploy

This tells Viewkit that testdeploy will ingest Log objects with the specified fields.

Check:

viewkit view inspect testdeploy
# now the query is attached to the view

6.4 Add SDL (GraphQL schema)

Add SDL to describe how the data is modeled/exposed:

viewkit view add sdl \
  "type FilteredAndDecodedLogs @materialized(if: false) {transactionHash: String}" \
  --name testdeploy

Notes:

@materialized(if: false)
Treat this as a virtual type, not a persisted table.
transactionHash: String
Minimal example field; real views will define more fields.

Inspect again:

viewkit view inspect testdeploy
# now shows both query and SDL

6.5 Attach a lens (WASM transform)

Attach a WebAssembly lens that filters logs by a specific address:

viewkit view add lens \
  --args '{"src":"address", "value":"0x1e3aA9fE4Ef01D3cB3189c129a49E3C03126C636"}' \
  --label "filter" \
  --url "https://raw.githubusercontent.com/shinzonetwork/wasm-bucket/main/bucket/filter_transaction/filter_transaction.wasm" \
  --name testdeploy

Flags:

--args – JSON passed to the lens
--label "filter" – human-readable label for the lens
--url – remote URL of the .wasm binary
--name testdeploy – attaches this lens to the testdeploy view

Inspect:

viewkit view inspect testdeploy
# you should now see:
# - query
# - SDL
# - lens "filter"

If you see libwasmer.dylib / “image not found” errors, revisit the Wasmer setup.

7. Wallet: create a deployment key

You need a wallet to sign deployments to local and devnet.

Generate one:

viewkit wallet generate

This will:

Create a new keypair
Store it in the wallet directory used by Viewkit
Print details such as the address (and possibly a mnemonic/seed)

Treat it like any other wallet:

Do not commit it to Git
Do not paste the mnemonic in public places
Store it securely

8. Deploy locally with `--target local`

The recommended flow is to deploy locally first, verify the view in a Playground, then deploy to a shared network like devnet.

Deploy to local:

viewkit view deploy testdeploy --target local

On success, you’ll see output similar to:

🚀 DefraDB is running on port 9181
⏳ Waiting for DefraDB to boot up...
✅ DefraDB booted up
⏳ Applying Schemas ...
✅ Schema Applied
⏳ Data Inserting...
✅ Data Inserted Successfully
✅ Applying View ...
✅ View Successfully Applied
🧪 Visit the DefraDB GraphQL Playground at http://127.0.0.1:9181/
📦 Press Ctrl+C to stop...

What’s happening:

A local DefraDB instance is started (port shown in the logs).
Schemas for your view are applied.
Any seed data (if configured) is inserted.
The view is applied.
A DefraDB GraphQL Playground URL is printed.

8.1 Use the DefraDB GraphQL Playground

Open the displayed URL in your browser, e.g.:
- http://127.0.0.1:9181/
You should see a GraphQL Playground / explorer.
You can:
- Inspect the schema (e.g. see FilteredAndDecodedLogs).
- Run test queries against your local view.
- Verify that your lens is filtering logs as expected.

Example query (adjust to your SDL):

{
  filteredAndDecodedLogs {
    transactionHash
  }
}

While this process is running, viewkit will keep the local DefraDB instance alive. To stop it:

Go back to the terminal and press Ctrl+C.

To iterate:

Update your view (queries/SDL/lenses).

Re-run:

viewkit view deploy testdeploy --target local

Refresh the Playground and test again.

9. Deploy to devnet with `--target devnet`

Once your view behaves correctly locally, you can deploy it to devnet.

Make sure:

A wallet has been generated: viewkit wallet generate
Any required devnet config/credentials are set

Then:

viewkit view deploy testdeploy --target devnet --rpc http://shinzohub-rpc.infra.source.network:8545

Conceptually:

Viewkit bundles the testdeploy view definition (queries, SDL, lenses, metadata).
Signs a deployment transaction using your wallet.
Sends it to the devnet network.
Registers the view on devnet so it can be used and queried there.

On success, you should see:

A transaction hash or deployment ID
A status message indicating success

If it fails:

Confirm your wallet is present and funded (if required)
Confirm devnet is a valid target
Re-check the view definition with viewkit view inspect testdeploy

10. Full flow cheat sheet

Here is the entire flow summarized:

# 0) clone + build
cd ~/code
git clone https://github.com/shinzonetwork/view-creator.git
cd view-creator
make build
export PATH="$PWD/build:$PATH"

# 1) initialize the view bundle
viewkit view init testdeploy

# 2) inspect the bundle
viewkit view inspect testdeploy

# 3) add a query (raw event shape to ingest)
viewkit view add query \
  "Log {address topics data transactionHash blockNumber}" \
  --name testdeploy

# 4) add SDL (how data is modeled/stored)
viewkit view add sdl \
  "type FilteredAndDecodedLogs @materialized(if: false) {transactionHash: String}" \
  --name testdeploy

# 5) attach a lens (WASM transform to filter by address)
viewkit view add lens \
  --args '{"src":"address", "value":"0x1e3aA9fE4Ef01D3cB3189c129a49E3C03126C636"}' \
  --label "filter" \
  --url "https://raw.githubusercontent.com/shinzonetwork/wasm-bucket/main/bucket/filter_transaction/filter_transaction.wasm" \
  --name testdeploy

# 6) create a wallet for deployments (one-time)
viewkit wallet generate

# 7) deploy locally and use the DefraDB Playground
viewkit view deploy testdeploy --target local
# -> follow the printed URL (e.g. http://127.0.0.1:9181/) for the GraphQL Playground
# -> press Ctrl+C in the terminal to stop

# 8) once you're happy, deploy to devnet
viewkit view deploy testdeploy --target devnet

This gives you a clean path from GitHub clone to a locally tested view and then to a devnet deployment.

1. Prerequisites​

2. Clone the repository​

3. Build the viewkit binary​

Optional: add viewkit to your PATH​

4. Wasmer runtime​

4.1 Install the Wasmer Go module​

4.2 Environment variables​

4.3 Configure the env vars​

5. Concepts: ViewsView A curated representation of defraDB documents defined using SDL. Views transform raw data into structured outputs., queries, SDL, lenses, wallets​

6. Quickstart: create and deploy a viewView A curated representation of defraDB documents defined using SDL. Views transform raw data into structured outputs. (testdeploy)​

6.1 Initialize the viewView A curated representation of defraDB documents defined using SDL. Views transform raw data into structured outputs. bundle​

6.2 Inspect the bundle​

6.3 Add a query (raw ingest shape)​

6.4 Add SDL (GraphQLGraphQL A query language and runtime used for retrieving structured data from defraDB, collections, and views within Shinzo. schema)​

6.5 Attach a lens (WASM transform)​

7. Wallet: create a deployment key​

8. Deploy locally with --target local​

8.1 Use the DefraDBdefraDB A NoSQL, document-oriented, edge-first database used by Shinzo to store encrypted or schema-defined data. GraphQLGraphQL A query language and runtime used for retrieving structured data from defraDB, collections, and views within Shinzo. Playground​

9. Deploy to devnet with --target devnet​

10. Full flow cheat sheet​