Add TaxonomyLinter to validate taxonomy TOML files against schema

.github/workflows/ci.yml

@@ -33,6 +33,9 @@ jobs:
       - name: Build
         run: cargo build --workspace --all-targets --all-features
 
+      - name: Lint taxonomy TOML
+        run: bash scripts/lint_taxonomy.sh
+
       - name: Run Tests
         run: cargo test --workspace --all-targets --all-features
 

Cargo.toml

@@ -98,5 +98,8 @@ tracing-subscriber = { version = "0.3", features = ["env-filter"] }
 # Metrics / Observability
 prometheus = "0.13"
 
+# URL parsing
+url = "2"
+
 # Shared internal crates
 grat-core = { path = "crates/core" }

crates/core/Cargo.toml

@@ -52,8 +52,16 @@ prometheus = { workspace = true }
 anyhow = { workspace = true }
 thiserror = { workspace = true }
 
+# URL parsing (for taxonomy linter documentation_url validation)
+url = { workspace = true }
+
+[[bin]]
+name = "taxonomy_linter"
+path = "src/bin/taxonomy_linter.rs"
+
 [dev-dependencies]
 tokio-test = "0.4"
 insta = { version = "1", features = ["json"] }
 tracing-subscriber = { workspace = true }
 ctor = "0.2"
+tempfile = "3"

crates/core/src/bin/taxonomy_linter.rs

@@ -0,0 +1,50 @@
+//! Taxonomy Linter — validate taxonomy TOML files against the schema.
+//!
+//! Usage:
+//!   taxonomy_linter [<directory>]
+//!
+//! Defaults to `crates/core/src/taxonomy/data` when no argument is given.
+//! Exits with status 0 if no issues found, 1 otherwise.
+
+use std::path::PathBuf;
+use std::process;
+
+fn main() {
+    let dir: PathBuf = std::env::args()
+        .nth(1)
+        .map(PathBuf::from)
+        .unwrap_or_else(|| {
+            // Default to the data directory relative to the binary's workspace root.
+            let mut p = PathBuf::from(env!("CARGO_MANIFEST_DIR"));
+            p.push("src/taxonomy/data");
+            p
+        });
+
+    if !dir.is_dir() {
+        eprintln!("error: '{}' is not a directory", dir.display());
+        process::exit(1);
+    }
+
+    let issues = match grat_core::taxonomy::linter::lint_dir(&dir) {
+        Ok(issues) => issues,
+        Err(e) => {
+            eprintln!("error: lint_dir failed: {e}");
+            process::exit(1);
+        }
+    };
+
+    if issues.is_empty() {
+        println!("✅ No issues found in {}", dir.display());
+        process::exit(0);
+    }
+
+    for issue in &issues {
+        match &issue.entry_id {
+            Some(eid) => eprintln!("{}:{}: {}", issue.file, eid, issue.message),
+            None => eprintln!("{}:: {}", issue.file, issue.message),
+        }
+    }
+
+    eprintln!("❌ Found {} issue(s) in {}", issues.len(), dir.display());
+    process::exit(1);
+}