enviodev · DZakh · May 22, 2026 · May 22, 2026 · May 25, 2026 · May 25, 2026
diff --git a/.github/workflows/build_and_verify.yml b/.github/workflows/build_and_verify.yml
@@ -156,7 +156,7 @@ jobs:
       - uses: ./.github/actions/cargo-cache
 
       - name: Hypersync Health Test
-        run: cargo test --features integration_tests
+        run: cargo test --features hypersync_health
 
   # Template tests - verify envio templates can be initialized and built
   template-tests:

diff --git a/CLAUDE.md b/CLAUDE.md
@@ -1,5 +1,6 @@
 - Use `pnpm` over `npm`/`npx`.
 - Always use single assert to check the whole value instead of multiple asserts for every field.
+- Never use `contains` in test assertions — assert the full expected value. Use `assert_eq!` or `insta::assert_snapshot!` for error messages.
 
 ## Comments
 

diff --git a/Cargo.lock b/Cargo.lock
diff --git a/packages/cli/Cargo.toml b/packages/cli/Cargo.toml
@@ -23,10 +23,11 @@ pathdiff = "0.2.3"
 serde = { version = "1.0.228", features = ["derive"] }
 serde_json = { version = "1.0.149", features = ["raw_value"] }
 serde_yaml = "0.9.19"
+json5 = "0.4"
 regex = "1.12"
 reqwest = { version = "0.11", features = ["json"] }
-strum = { version = "0.26", features = ["derive"] }
-strum_macros = "0.26"
+strum = { version = "0.27", features = ["derive"] }
+strum_macros = "0.27"
 tokio = { version = "1.49", features = [
   "macros",
   "process",
@@ -45,6 +46,7 @@ colored = "2.0.4"
 thiserror = "1.0.50"
 fuel-abi-types = "0.7.0"
 hypersync-client = "1.2.0"
+arrow = { version = "57", default-features = false, features = ["prettyprint"] }
 faster-hex = "0.9"
 ruint = "1"
 env_logger = "0.11"
@@ -59,7 +61,12 @@ napi = { version = "=3.8.5", features = ["napi10", "async", "serde-json"] }
 napi-derive = "=3.5.4"
 
 [features]
-integration_tests = []
+hypersync_health = []
+# Compatibility alias for the previous feature name. Older CI workflows
+# (still on main until this PR merges) reference `integration_tests`;
+# enabling it must keep the hypersync health check working. Drop after
+# main picks up the renamed workflow step.
+integration_tests = ["hypersync_health"]
 
 [build-dependencies]
 napi-build = "=2.3.1"

diff --git a/packages/cli/CommandLineHelp.md b/packages/cli/CommandLineHelp.md
@@ -29,6 +29,7 @@ This document contains the help content for the `envio` command-line program.
 * [`envio local db-migrate setup`↴](#envio-local-db-migrate-setup)
 * [`envio start`↴](#envio-start)
 * [`envio metrics`↴](#envio-metrics)
+* [`envio data`↴](#envio-data)
 * [`envio skills`↴](#envio-skills)
 * [`envio skills update`↴](#envio-skills-update)
 * [`envio tools`↴](#envio-tools)
@@ -50,6 +51,7 @@ This document contains the help content for the `envio` command-line program.
 * `local` — Prepare local environment for envio testing
 * `start` — Start the indexer. Runs codegen automatically before launching so the on-disk types stay in sync with `config.yaml` and `schema.graphql`
 * `metrics` — Fetch raw Prometheus metrics from the running indexer's /metrics endpoint
+* `data` — Query raw blockchain data — blocks, logs, transactions on EVM chains using the same `where` syntax as indexer filters
 * `skills` — Manage Envio-provided Claude Code skills under `.claude/skills/`
 * `tools` — Tools for people and AI agents (search-docs, fetch-docs). Run `envio tools help` for details
 * `config` — Inspect the indexer config
@@ -381,6 +383,29 @@ Fetch raw Prometheus metrics from the running indexer's /metrics endpoint
 
 
 
+## `envio data`
+
+Query raw blockchain data — blocks, logs, transactions on EVM chains using the same `where` syntax as indexer filters.
+
+Output is TOON (token-oriented) tabular form.
+
+Example — earliest USDC transfers on Base: `envio data block.number log.srcAddress --chain=base --where='{ block: { number: { _gte: 0 } }, log: { srcAddress: "0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913" } }'`
+
+Example — current archive height: `envio data knownHeight --chain=arbitrum-one`
+
+**Usage:** `envio data [OPTIONS] --chain <CHAIN> [FIELD]...`
+
+###### **Arguments:**
+
+* `<FIELD>` — Fields to fetch (e.g. `block.number`, `log.srcAddress`, `transaction.transactionIndex`). Use `knownHeight` to get the chain's current archive height
+
+###### **Options:**
+
+* `--chain <CHAIN>` — Chain id (e.g. `8453`) or kebab-case name (e.g. `base`, `arbitrum-one`). Solana is not supported yet
+* `--where <WHERE_FILTER>` — Filter rows (JSON5: unquoted keys, single quotes, trailing commas, `//` comments). Group fields under `block`, `transaction`, `log`. Match any field with a value, array, or `_eq`/`_in`; numeric fields also take `_gt`/`_gte`/`_lt`/`_lte`. Example: --where='{ block: { number: { _gte: 1000, _lte: 2000 } }, log: { srcAddress: "0xa0b8..." } }'
+
+
+
 ## `envio skills`
 
 Manage Envio-provided Claude Code skills under `.claude/skills/`

diff --git a/packages/cli/examples/script.rs b/packages/cli/examples/script.rs
@@ -22,13 +22,16 @@ async fn main() -> Result<()> {
     )
     .context("Failed parsing command line arguments")?;
 
-    // Guard against accidental misuse — only `script` subcommands go through
-    // this path. Everything else (init/codegen/dev/start/...) runs via the
-    // NAPI host and would be missing the JS dispatch side if invoked here.
-    if !matches!(command_line_args.command, CommandType::Script(_)) {
+    // Guard against accidental misuse — only commands that finish entirely
+    // in Rust (no JS dispatch needed) go through this path. Everything else
+    // (init/codegen/dev/start/...) runs via the NAPI host.
+    if !matches!(
+        command_line_args.command,
+        CommandType::Script(_) | CommandType::Data(_)
+    ) {
         anyhow::bail!(
-            "This example only supports `script` subcommands. Run envio via the NAPI host \
-             (packages/envio/bin.mjs) for init/codegen/dev/start/etc."
+            "This example only supports `script` and `data` subcommands. Run envio via the NAPI \
+             host (packages/envio/bin.mjs) for init/codegen/dev/start/etc."
         );
     }
 

diff --git a/packages/cli/src/cli_args/clap_definitions.rs b/packages/cli/src/cli_args/clap_definitions.rs
@@ -61,6 +61,18 @@ pub enum CommandType {
     ///Fetch raw Prometheus metrics from the running indexer's /metrics endpoint
     Metrics,
 
+    ///Query raw blockchain data — blocks, logs, transactions on EVM chains
+    ///using the same `where` syntax as indexer filters.
+    ///
+    ///Output is TOON (token-oriented) tabular form.
+    ///
+    ///Example — earliest USDC transfers on Base:
+    ///`envio data block.number log.srcAddress --chain=base --where='{ block: { number: { _gte: 0 } }, log: { srcAddress: "0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913" } }'`
+    ///
+    ///Example — current archive height:
+    ///`envio data knownHeight --chain=arbitrum-one`
+    Data(DataArgs),
+
     ///Manage Envio-provided Claude Code skills under `.claude/skills/`
     #[command(subcommand)]
     Skills(SkillsSubcommand),
@@ -130,6 +142,28 @@ pub enum JsonSchema {
     Svm,
 }
 
+#[derive(Debug, Args)]
+pub struct DataArgs {
+    ///Fields to fetch (e.g. `block.number`, `log.srcAddress`,
+    ///`transaction.transactionIndex`). Use `knownHeight` to get the
+    ///chain's current archive height.
+    #[arg(value_name = "FIELD")]
+    pub fields: Vec<String>,
+
+    ///Chain id (e.g. `8453`) or kebab-case name (e.g. `base`,
+    ///`arbitrum-one`). Solana is not supported yet.
+    #[arg(long)]
+    pub chain: String,
+
+    ///Filter rows (JSON5: unquoted keys, single quotes, trailing commas,
+    ///`//` comments). Group fields under `block`, `transaction`, `log`.
+    ///Match any field with a value, array, or `_eq`/`_in`; numeric fields
+    ///also take `_gt`/`_gte`/`_lt`/`_lte`. Example:
+    ///--where='{ block: { number: { _gte: 1000, _lte: 2000 } }, log: { srcAddress: "0xa0b8..." } }'
+    #[arg(long = "where")]
+    pub where_filter: Option<String>,
+}
+
 #[derive(Debug, Args)]
 pub struct DevArgs {
     ///Force restart: clear the database and re-index from scratch. Required when config/schema/ABI changes are incompatible with the existing indexer state.

diff --git a/.../src/cli_args/snapshots/envio__cli_args__clap_definitions__test__envio_help_snapshot.snap b/.../src/cli_args/snapshots/envio__cli_args__clap_definitions__test__envio_help_snapshot.snap
@@ -12,6 +12,7 @@ Commands:
   local    Prepare local environment for envio testing
   start    Start the indexer. Runs codegen automatically before launching so the on-disk types stay in sync with `config.yaml` and `schema.graphql`
   metrics  Fetch raw Prometheus metrics from the running indexer's /metrics endpoint
+  data     Query raw blockchain data — blocks, logs, transactions on EVM chains using the same `where` syntax as indexer filters
   skills   Manage Envio-provided Claude Code skills under `.claude/skills/`
   tools    Tools for people and AI agents (search-docs, fetch-docs). Run `envio tools help` for details
   config   Inspect the indexer config

diff --git a/packages/cli/src/config_parsing/hypersync_endpoints.rs b/packages/cli/src/config_parsing/hypersync_endpoints.rs
@@ -34,11 +34,11 @@ mod test {
     }
 }
 
-/// Integration tests that require network access.
-/// Run with: cargo test --features integration_tests
+/// Hypersync endpoint health checks that require network access.
+/// Run with: cargo test --features hypersync_health
 #[cfg(test)]
-#[cfg(feature = "integration_tests")]
-mod integration_tests {
+#[cfg(feature = "hypersync_health")]
+mod hypersync_health {
     use super::{network_to_hypersync_url, HypersyncChain};
     use crate::scripts::print_missing_networks::Diff;
     use strum::IntoEnumIterator;