diff --git a/backend/Cargo.lock b/backend/Cargo.lock
index 3e3319a..ab6c3e8 100644
--- a/backend/Cargo.lock
+++ b/backend/Cargo.lock
@@ -462,6 +462,28 @@ dependencies = [
  "windows-link 0.2.1",
 ]
 
+[[package]]
+name = "chrono-tz"
+version = "0.8.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d59ae0466b83e838b81a54256c39d5d7c20b9d7daa10510a242d9b75abd5936e"
+dependencies = [
+ "chrono",
+ "chrono-tz-build",
+ "phf",
+]
+
+[[package]]
+name = "chrono-tz-build"
+version = "0.2.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "433e39f13c9a060046954e0592a8d0a4bcb1040125cbf91cb8ee58964cfb350f"
+dependencies = [
+ "parse-zoneinfo",
+ "phf",
+ "phf_codegen",
+]
+
 [[package]]
 name = "clang-sys"
 version = "1.8.1"
@@ -2100,6 +2122,7 @@ dependencies = [
  "base64 0.22.1",
  "bytes",
  "chrono",
+ "chrono-tz",
  "clap",
  "diesel",
  "diesel_migrations",
@@ -2117,6 +2140,7 @@ dependencies = [
  "percent-encoding",
  "quick-xml",
  "rand 0.9.2",
+ "regex",
  "reqwest",
  "rust-s3",
  "serde",
@@ -2124,6 +2148,7 @@ dependencies = [
  "serde_bytes",
  "serde_cbor_2",
  "serde_json",
+ "serde_yaml",
  "sha2",
  "tempfile",
  "thiserror 2.0.17",
@@ -2162,6 +2187,15 @@ dependencies = [
  "windows-link 0.2.1",
 ]
 
+[[package]]
+name = "parse-zoneinfo"
+version = "0.3.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1f2a05b18d44e2957b88f96ba460715e295bc1d7510468a2f3d3b44535d26c24"
+dependencies = [
+ "regex",
+]
+
 [[package]]
 name = "password-hash"
 version = "0.5.0"
@@ -2230,6 +2264,44 @@ version = "2.3.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "9b4f627cb1b25917193a259e49bdad08f671f8d9708acfd5fe0a8c1455d87220"
 
+[[package]]
+name = "phf"
+version = "0.11.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1fd6780a80ae0c52cc120a26a1a42c1ae51b247a253e4e06113d23d2c2edd078"
+dependencies = [
+ "phf_shared",
+]
+
+[[package]]
+name = "phf_codegen"
+version = "0.11.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "aef8048c789fa5e851558d709946d6d79a8ff88c0440c587967f8e94bfb1216a"
+dependencies = [
+ "phf_generator",
+ "phf_shared",
+]
+
+[[package]]
+name = "phf_generator"
+version = "0.11.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3c80231409c20246a13fddb31776fb942c38553c51e871f8cbd687a4cfb5843d"
+dependencies = [
+ "phf_shared",
+ "rand 0.8.5",
+]
+
+[[package]]
+name = "phf_shared"
+version = "0.11.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "67eabc2ef2a60eb7faa00097bd1ffdb5bd28e62bf39990626a582201b7a754e5"
+dependencies = [
+ "siphasher",
+]
+
 [[package]]
 name = "pin-project-lite"
 version = "0.2.16"
@@ -3001,6 +3073,19 @@ dependencies = [
  "serde",
 ]
 
+[[package]]
+name = "serde_yaml"
+version = "0.9.34+deprecated"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6a8b1a1a2ebf674015cc02edccce75287f1a0130d394307b36743c2f5d504b47"
+dependencies = [
+ "indexmap",
+ "itoa",
+ "ryu",
+ "serde",
+ "unsafe-libyaml",
+]
+
 [[package]]
 name = "sha1"
 version = "0.10.6"
@@ -3075,6 +3160,12 @@ dependencies = [
  "time",
 ]
 
+[[package]]
+name = "siphasher"
+version = "1.0.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "56199f7ddabf13fe5074ce809e7d3f42b42ae711800501b5b16ea82ad029c39d"
+
 [[package]]
 name = "slab"
 version = "0.4.11"
@@ -3555,6 +3646,12 @@ version = "1.0.22"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "9312f7c4f6ff9069b165498234ce8be658059c6728633667c526e27dc2cf1df5"
 
+[[package]]
+name = "unsafe-libyaml"
+version = "0.2.11"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "673aac59facbab8a9007c7f6108d11f63b603f7cabff99fabf650fea5c32b861"
+
 [[package]]
 name = "untrusted"
 version = "0.9.0"
diff --git a/backend/Cargo.toml b/backend/Cargo.toml
index dddccc8..f4ce6ce 100644
--- a/backend/Cargo.toml
+++ b/backend/Cargo.toml
@@ -45,6 +45,7 @@ quick-xml = "0.38"
 futures-util = "0.3"
 url = "2.5"
 once_cell = "1.19"
+regex = "1.11"
 utoipa = { version = "4.2", default-features = false, features = ["chrono", "uuid", "preserve_order"] }
 clap = { version = "4.5", features = ["derive"] }
 
@@ -63,10 +64,18 @@ serde_cbor_2 = "0.13"
 rand = "0.9"
 hyper = "1.2"
 http-body-util = "0.1"
+chrono-tz = "0.8"
 
 [dev-dependencies]
 once_cell = "1.19"
 webauthn-rs-core = "0.5"
+serde_yaml = "0.9"
+
+[build-dependencies]
+serde = { version = "1.0", features = ["derive"] }
+serde_yaml = "0.9"
+serde_json = "1.0"
+regex = "1.11"
 
 [[bin]]
 name = "backend"
diff --git a/backend/build.rs b/backend/build.rs
new file mode 100644
index 0000000..e040bf0
--- /dev/null
+++ b/backend/build.rs
@@ -0,0 +1,123 @@
+use std::env;
+use std::fs;
+use std::path::PathBuf;
+
+use regex::escape;
+use serde::Deserialize;
+
+#[derive(Deserialize)]
+struct CaseSuite {
+    cases: Vec<CaseName>,
+}
+
+#[derive(Deserialize)]
+struct CaseName {
+    name: String,
+}
+
+#[derive(Deserialize)]
+struct MonthSuite {
+    months: Vec<MonthDefinition>,
+}
+
+#[derive(Deserialize)]
+struct MonthDefinition {
+    name: String,
+    month: u32,
+    #[serde(default)]
+    locales: Vec<String>,
+}
+
+fn sanitize(name: &str) -> String {
+    let mut out = String::with_capacity(name.len());
+    for ch in name.chars() {
+        if ch.is_ascii_alphanumeric() {
+            out.push(ch);
+        } else {
+            out.push('_');
+        }
+    }
+    if out.is_empty() {
+        "case".to_string()
+    } else if out.chars().next().unwrap().is_ascii_digit() {
+        format!("_{}", out)
+    } else {
+        out
+    }
+}
+
+fn quote(value: &str) -> String {
+    serde_json::to_string(value).expect("string literal")
+}
+
+fn generate_tests() -> Result<String, Box<dyn std::error::Error>> {
+    let yaml_path = PathBuf::from("tests/data/issued_at_cases.yaml");
+    let contents = fs::read_to_string(&yaml_path)?;
+    let suite: CaseSuite = serde_yaml::from_str(&contents)?;
+
+    let mut output =
+        String::from("#[cfg(test)]\npub mod issued_at_generated_tests {\n    use super::*;\n");
+
+    for case in suite.cases {
+        let ident = sanitize(&case.name);
+        output.push_str(&format!(
+            "    #[test]\n    fn {}() {{\n        run_named_case(\"{}\");\n    }}\n",
+            ident, case.name
+        ));
+    }
+
+    output.push_str("}\n");
+    Ok(output)
+}
+
+fn generate_months() -> Result<String, Box<dyn std::error::Error>> {
+    let yaml_path = PathBuf::from("resources/issued_at_months.yaml");
+    let contents = fs::read_to_string(&yaml_path)?;
+    let suite: MonthSuite = serde_yaml::from_str(&contents)?;
+
+    let mut pattern_parts = Vec::with_capacity(suite.months.len());
+    let mut entries = String::new();
+    for entry in &suite.months {
+        pattern_parts.push(escape(&entry.name));
+        let locales_literal = if entry.locales.is_empty() {
+            "&[]".to_string()
+        } else {
+            let joined = entry
+                .locales
+                .iter()
+                .map(|loc| quote(loc))
+                .collect::<Vec<_>>()
+                .join(", ");
+            format!("&[{}]", joined)
+        };
+        entries.push_str(&format!(
+            "    MonthVariant {{ name: {}, month: {}, locales: {} }},\n",
+            quote(&entry.name),
+            entry.month,
+            locales_literal
+        ));
+    }
+
+    let pattern_literal = quote(&pattern_parts.join("|"));
+    let output = format!(
+        "pub(super) static MONTH_VARIANTS: &[MonthVariant] = &[\n{entries}];\n\n",
+        entries = entries
+    ) + &format!(
+        "pub(super) const MONTH_PATTERN: &str = {};\n",
+        pattern_literal
+    );
+    Ok(output)
+}
+
+fn main() -> Result<(), Box<dyn std::error::Error>> {
+    println!("cargo:rerun-if-changed=tests/data/issued_at_cases.yaml");
+    println!("cargo:rerun-if-changed=resources/issued_at_months.yaml");
+
+    let out_dir = PathBuf::from(env::var("OUT_DIR")?);
+    fs::write(
+        out_dir.join("issued_at_generated_tests.rs"),
+        generate_tests()?,
+    )?;
+    fs::write(out_dir.join("issued_at_months.rs"), generate_months()?)?;
+    Ok(())
+}
diff --git a/backend/resources/issued_at_months.yaml b/backend/resources/issued_at_months.yaml
new file mode 100644
index 0000000..28c727d
--- /dev/null
+++ b/backend/resources/issued_at_months.yaml
@@ -0,0 +1,648 @@
+months:
+  - name: "january"
+    month: 1
+    locales:
+      - "en"
+  - name: "jan"
+    month: 1
+    locales:
+      - "en"
+  - name: "janvier"
+    month: 1
+    locales:
+      - "fr"
+  - name: "janv"
+    month: 1
+    locales:
+      - "fr"
+  - name: "januar"
+    month: 1
+    locales:
+      - "de"
+  - name: "janu\u00e1r"
+    month: 1
+    locales:
+      - "hu"
+  - name: "leden"
+    month: 1
+    locales:
+      - "cs"
+  - name: "sije\u010danj"
+    month: 1
+    locales:
+      - "hr"
+  - name: "sijecanj"
+    month: 1
+    locales:
+      - "hr"
+  - name: "stycze\u0144"
+    month: 1
+    locales:
+      - "pl"
+  - name: "styczen"
+    month: 1
+    locales:
+      - "pl"
+  - name: "ocak"
+    month: 1
+    locales:
+      - "tr"
+  - name: "february"
+    month: 2
+    locales:
+      - "en"
+  - name: "feb"
+    month: 2
+    locales:
+      - "en"
+  - name: "f\u00e9vrier"
+    month: 2
+    locales:
+      - "fr"
+  - name: "fevrier"
+    month: 2
+    locales:
+      - "fr"
+  - name: "f\u00e9vr"
+    month: 2
+    locales:
+      - "fr"
+  - name: "fevr"
+    month: 2
+    locales:
+      - "fr"
+  - name: "februar"
+    month: 2
+    locales:
+      - "de"
+  - name: "\u00fanor"
+    month: 2
+    locales:
+      - "cs"
+  - name: "unor"
+    month: 2
+    locales:
+      - "cs"
+  - name: "\u00fanora"
+    month: 2
+    locales:
+      - "cs"
+  - name: "unora"
+    month: 2
+    locales:
+      - "cs"
+  - name: "velja\u010da"
+    month: 2
+    locales:
+      - "hr"
+  - name: "veljaca"
+    month: 2
+    locales:
+      - "hr"
+  - name: "velja\u010de"
+    month: 2
+    locales:
+      - "hr"
+  - name: "veljace"
+    month: 2
+    locales:
+      - "hr"
+  - name: "luty"
+    month: 2
+    locales:
+      - "pl"
+  - name: "\u015fubat"
+    month: 2
+    locales:
+      - "tr"
+  - name: "subat"
+    month: 2
+    locales:
+      - "tr"
+  - name: "march"
+    month: 3
+    locales:
+      - "en"
+  - name: "mar"
+    month: 3
+    locales:
+      - "en"
+  - name: "m\u00e4rz"
+    month: 3
+    locales:
+      - "de"
+  - name: "maerz"
+    month: 3
+    locales:
+      - "de"
+  - name: "mars"
+    month: 3
+    locales:
+      - "fr"
+  - name: "m\u00e4r"
+    month: 3
+    locales:
+      - "de"
+  - name: "marz"
+    month: 3
+    locales:
+      - "de"
+  - name: "b\u0159ezen"
+    month: 3
+    locales:
+      - "cs"
+  - name: "brezen"
+    month: 3
+    locales:
+      - "cs"
+  - name: "b\u0159ezna"
+    month: 3
+    locales:
+      - "cs"
+  - name: "brezna"
+    month: 3
+    locales:
+      - "cs"
+  - name: "o\u017eujak"
+    month: 3
+    locales:
+      - "hr"
+  - name: "ozujak"
+    month: 3
+    locales:
+      - "hr"
+  - name: "o\u017eujka"
+    month: 3
+    locales:
+      - "hr"
+  - name: "ozujka"
+    month: 3
+    locales:
+      - "hr"
+  - name: "marzec"
+    month: 3
+    locales:
+      - "pl"
+  - name: "mart"
+    month: 3
+    locales:
+      - "sr"
+      - "bs"
+  - name: "m\u00e1rcius"
+    month: 3
+    locales:
+      - "hu"
+  - name: "martie"
+    month: 3
+    locales:
+      - "ro"
+  - name: "april"
+    month: 4
+    locales:
+      - "en"
+  - name: "apr"
+    month: 4
+    locales:
+      - "en"
+  - name: "avril"
+    month: 4
+    locales:
+      - "fr"
+  - name: "abril"
+    month: 4
+    locales:
+      - "es"
+      - "pt"
+  - name: "duben"
+    month: 4
+    locales:
+      - "cs"
+  - name: "travanj"
+    month: 4
+    locales:
+      - "hr"
+  - name: "nisan"
+    month: 4
+    locales:
+      - "tr"
+  - name: "kwiecie\u0144"
+    month: 4
+    locales:
+      - "pl"
+  - name: "kwiecien"
+    month: 4
+    locales:
+      - "pl"
+  - name: "aprile"
+    month: 4
+    locales:
+      - "it"
+  - name: "may"
+    month: 5
+    locales:
+      - "en"
+  - name: "mai"
+    month: 5
+    locales:
+      - "fr"
+      - "de"
+  - name: "mayo"
+    month: 5
+    locales:
+      - "es"
+  - name: "kv\u011bten"
+    month: 5
+    locales:
+      - "cs"
+  - name: "kveten"
+    month: 5
+    locales:
+      - "cs"
+  - name: "kv\u011btna"
+    month: 5
+    locales:
+      - "cs"
+  - name: "kvetna"
+    month: 5
+    locales:
+      - "cs"
+  - name: "svibanj"
+    month: 5
+    locales:
+      - "hr"
+  - name: "maj"
+    month: 5
+    locales:
+      - "pl"
+      - "bs"
+      - "sr"
+      - "hr"
+  - name: "may\u0131s"
+    month: 5
+    locales:
+      - "tr"
+  - name: "mayis"
+    month: 5
+    locales:
+      - "tr"
+  - name: "maggio"
+    month: 5
+    locales:
+      - "it"
+  - name: "m\u00e1j"
+    month: 5
+    locales:
+      - "hu"
+      - "sk"
+  - name: "june"
+    month: 6
+    locales:
+      - "en"
+  - name: "jun"
+    month: 6
+    locales:
+      - "en"
+      - "de"
+  - name: "juin"
+    month: 6
+    locales:
+      - "fr"
+  - name: "junio"
+    month: 6
+    locales:
+      - "es"
+  - name: "juni"
+    month: 6
+    locales:
+      - "de"
+  - name: "\u010derven"
+    month: 6
+    locales:
+      - "cs"
+  - name: "cerven"
+    month: 6
+    locales:
+      - "cs"
+  - name: "lipanj"
+    month: 6
+    locales:
+      - "hr"
+  - name: "haziran"
+    month: 6
+    locales:
+      - "tr"
+  - name: "czerwiec"
+    month: 6
+    locales:
+      - "pl"
+  - name: "j\u00fanius"
+    month: 6
+    locales:
+      - "hu"
+  - name: "giugno"
+    month: 6
+    locales:
+      - "it"
+  - name: "july"
+    month: 7
+    locales:
+      - "en"
+  - name: "jul"
+    month: 7
+    locales:
+      - "en"
+      - "de"
+  - name: "juillet"
+    month: 7
+    locales:
+      - "fr"
+  - name: "julio"
+    month: 7
+    locales:
+      - "es"
+  - name: "temmuz"
+    month: 7
+    locales:
+      - "tr"
+  - name: "\u010dervenec"
+    month: 7
+    locales:
+      - "cs"
+  - name: "cervenec"
+    month: 7
+    locales:
+      - "cs"
+  - name: "srpanj"
+    month: 7
+    locales:
+      - "hr"
+  - name: "lipiec"
+    month: 7
+    locales:
+      - "pl"
+  - name: "luglio"
+    month: 7
+    locales:
+      - "it"
+  - name: "j\u00falius"
+    month: 7
+    locales:
+      - "hu"
+  - name: "august"
+    month: 8
+    locales:
+      - "en"
+  - name: "aug"
+    month: 8
+    locales:
+      - "en"
+      - "de"
+  - name: "ao\u00fbt"
+    month: 8
+    locales:
+      - "fr"
+  - name: "aout"
+    month: 8
+    locales:
+      - "fr"
+  - name: "agosto"
+    month: 8
+    locales:
+      - "es"
+      - "pt"
+      - "it"
+  - name: "a\u011fustos"
+    month: 8
+    locales:
+      - "tr"
+  - name: "agustos"
+    month: 8
+    locales:
+      - "tr"
+  - name: "kolovoz"
+    month: 8
+    locales:
+      - "hr"
+  - name: "srpen"
+    month: 8
+    locales:
+      - "cs"
+  - name: "sierpie\u0144"
+    month: 8
+    locales:
+      - "pl"
+  - name: "sierpien"
+    month: 8
+    locales:
+      - "pl"
+  - name: "augustus"
+    month: 8
+    locales:
+      - "nl"
+  - name: "agost"
+    month: 8
+    locales:
+      - "it"
+      - "ca"
+  - name: "september"
+    month: 9
+    locales:
+      - "en"
+  - name: "sept"
+    month: 9
+    locales:
+      - "en"
+      - "fr"
+  - name: "sep"
+    month: 9
+    locales:
+      - "en"
+  - name: "septembre"
+    month: 9
+    locales:
+      - "fr"
+  - name: "septiembre"
+    month: 9
+    locales:
+      - "es"
+  - name: "eyl\u00fcl"
+    month: 9
+    locales:
+      - "tr"
+  - name: "eylul"
+    month: 9
+    locales:
+      - "tr"
+  - name: "z\u00e1\u0159\u00ed"
+    month: 9
+    locales:
+      - "cs"
+  - name: "zari"
+    month: 9
+    locales:
+      - "cs"
+  - name: "rujan"
+    month: 9
+    locales:
+      - "hr"
+  - name: "wrzesie\u0144"
+    month: 9
+    locales:
+      - "pl"
+  - name: "wrzesien"
+    month: 9
+    locales:
+      - "pl"
+  - name: "septembrie"
+    month: 9
+    locales:
+      - "ro"
+  - name: "settembre"
+    month: 9
+    locales:
+      - "it"
+  - name: "october"
+    month: 10
+    locales:
+      - "en"
+  - name: "oct"
+    month: 10
+    locales:
+      - "en"
+  - name: "oktober"
+    month: 10
+    locales:
+      - "de"
+  - name: "okt\u00f3ber"
+    month: 10
+    locales:
+      - "hu"
+  - name: "octobre"
+    month: 10
+    locales:
+      - "fr"
+  - name: "octubre"
+    month: 10
+    locales:
+      - "es"
+  - name: "\u0159\u00edjen"
+    month: 10
+    locales:
+      - "cs"
+  - name: "rijen"
+    month: 10
+    locales:
+      - "hr"
+  - name: "pa\u017adziernik"
+    month: 10
+    locales:
+      - "pl"
+  - name: "pazdziernik"
+    month: 10
+    locales:
+      - "pl"
+  - name: "ekim"
+    month: 10
+    locales:
+      - "tr"
+  - name: "octombrie"
+    month: 10
+    locales:
+      - "ro"
+  - name: "november"
+    month: 11
+    locales:
+      - "en"
+  - name: "nov"
+    month: 11
+    locales:
+      - "en"
+  - name: "novembre"
+    month: 11
+    locales:
+      - "fr"
+  - name: "noviembre"
+    month: 11
+    locales:
+      - "es"
+  - name: "studeni"
+    month: 11
+    locales:
+      - "hr"
+  - name: "kas\u0131m"
+    month: 11
+    locales:
+      - "tr"
+  - name: "kasim"
+    month: 11
+    locales:
+      - "tr"
+  - name: "listopad"
+    month: 11
+    locales:
+      - "pl"
+      - "cs"
+  - name: "novembro"
+    month: 11
+    locales:
+      - "pt"
+  - name: "listopadu"
+    month: 11
+    locales:
+      - "cs"
+  - name: "noiembrie"
+    month: 11
+    locales:
+      - "ro"
+  - name: "december"
+    month: 12
+    locales:
+      - "en"
+  - name: "dec"
+    month: 12
+    locales:
+      - "en"
+  - name: "dezember"
+    month: 12
+    locales:
+      - "de"
+  - name: "d\u00e9cembre"
+    month: 12
+    locales:
+      - "fr"
+  - name: "decembre"
+    month: 12
+    locales:
+      - "fr"
+  - name: "prosinec"
+    month: 12
+    locales:
+      - "cs"
+  - name: "prosinac"
+    month: 12
+    locales:
+      - "hr"
+  - name: "grudzie\u0144"
+    month: 12
+    locales:
+      - "pl"
+  - name: "grudzien"
+    month: 12
+    locales:
+      - "pl"
+  - name: "grudnia"
+    month: 12
+    locales:
+      - "pl"
+  - name: "aral\u0131k"
+    month: 12
+    locales:
+      - "tr"
+  - name: "aralik"
+    month: 12
+    locales:
+      - "tr"
+  - name: "decembrie"
+    month: 12
+    locales:
+      - "ro"
diff --git a/backend/src/config.rs b/backend/src/config.rs
index d095a96..6fc2e77 100644
--- a/backend/src/config.rs
+++ b/backend/src/config.rs
@@ -1,6 +1,7 @@
 use anyhow::{Context, Result};
 use url::Url;
 
+use serde::de::Deserializer;
 use serde::Deserialize;
 use serde_aux::field_attributes::deserialize_bool_from_anything;
 
@@ -60,6 +61,16 @@ pub struct AppConfig {
     pub worker_max_document_bytes: u64,
     #[serde(default = "default_upload_body_limit_bytes")]
     pub upload_body_limit_bytes: u64,
+    #[serde(default = "default_service_timezone")]
+    pub service_timezone: String,
+    #[serde(default = "default_issued_at_date_order")]
+    pub issued_at_date_order: String,
+    #[serde(default)]
+    pub issued_at_filename_date_order: Option<String>,
+    #[serde(default, deserialize_with = "deserialize_string_list")]
+    pub issued_at_date_parser_locales: Vec<String>,
+    #[serde(default, deserialize_with = "deserialize_string_list")]
+    pub issued_at_ignore_dates: Vec<String>,
     #[serde(default)]
     pub webauthn_rp_id: Option<String>,
     #[serde(default)]
@@ -183,6 +194,52 @@ fn default_upload_body_limit_bytes() -> u64 {
     128 * 1024 * 1024
 }
 
+fn default_service_timezone() -> String {
+    "UTC".to_string()
+}
+
+fn default_issued_at_date_order() -> String {
+    "DMY".to_string()
+}
+
+fn deserialize_string_list<'de, D>(deserializer: D) -> Result<Vec<String>, D::Error>
+where
+    D: Deserializer<'de>,
+{
+    #[derive(Deserialize)]
+    #[serde(untagged)]
+    enum Helper {
+        List(Vec<String>),
+        Single(String),
+    }
+
+    let helper = Option::<Helper>::deserialize(deserializer)?;
+    let mut values = Vec::new();
+
+    if let Some(helper) = helper {
+        match helper {
+            Helper::List(list) => {
+                for entry in list {
+                    let trimmed = entry.trim();
+                    if !trimmed.is_empty() {
+                        values.push(trimmed.to_string());
+                    }
+                }
+            }
+            Helper::Single(value) => {
+                for part in value.split(',') {
+                    let trimmed = part.trim();
+                    if !trimmed.is_empty() {
+                        values.push(trimmed.to_string());
+                    }
+                }
+            }
+        }
+    }
+
+    Ok(values)
+}
+
 fn default_webauthn_rp_name() -> String {
     "Papercrate".to_string()
 }
diff --git a/backend/src/issued_at.rs b/backend/src/issued_at.rs
new file mode 100644
index 0000000..f70487e
--- /dev/null
+++ b/backend/src/issued_at.rs
@@ -0,0 +1,167 @@
+use std::collections::HashSet;
+
+use chrono::{DateTime, Datelike, NaiveDate, TimeZone, Utc};
+use chrono_tz::Tz;
+use once_cell::sync::Lazy;
+use tracing::warn;
+
+use crate::config::AppConfig;
+
+#[derive(Clone, Copy, Debug)]
+pub enum DateOrder {
+    Dmy,
+    Mdy,
+    Ymd,
+}
+
+impl DateOrder {
+    pub fn parse(value: &str) -> Self {
+        Self::try_parse(value).unwrap_or(DateOrder::Dmy)
+    }
+
+    pub fn try_parse(value: &str) -> Option<Self> {
+        match value.trim().to_ascii_uppercase().as_str() {
+            "YMD" => Some(DateOrder::Ymd),
+            "MDY" => Some(DateOrder::Mdy),
+            "DMY" => Some(DateOrder::Dmy),
+            _ => None,
+        }
+    }
+}
+
+static MIN_ISSUED_AT_DATE: Lazy<NaiveDate> =
+    Lazy::new(|| NaiveDate::from_ymd_opt(1901, 1, 1).expect("valid minimum issued_at date"));
+
+#[derive(Clone, Debug)]
+pub struct IssuedAtSettings {
+    pub timezone: Tz,
+    pub date_order: DateOrder,
+    pub filename_date_order: Option<DateOrder>,
+    pub locales: HashSet<String>,
+    pub ignore_dates: HashSet<NaiveDate>,
+    pub min_date: NaiveDate,
+}
+
+impl IssuedAtSettings {
+    pub fn from_config(config: &AppConfig) -> Self {
+        let timezone = config.service_timezone.parse::<Tz>().unwrap_or_else(|_| {
+            warn!(
+                timezone = %config.service_timezone,
+                "invalid service timezone configured; falling back to UTC"
+            );
+            chrono_tz::UTC
+        });
+
+        let date_order = DateOrder::parse(&config.issued_at_date_order);
+        let filename_date_order =
+            config
+                .issued_at_filename_date_order
+                .as_deref()
+                .and_then(|value| {
+                    DateOrder::try_parse(value).or_else(|| {
+                        warn!(value, "invalid issued_at filename date order; ignoring");
+                        None
+                    })
+                });
+
+        let locales = config
+            .issued_at_date_parser_locales
+            .iter()
+            .filter_map(|value| {
+                let trimmed = value.trim();
+                if trimmed.is_empty() {
+                    None
+                } else {
+                    Some(trimmed.to_ascii_lowercase())
+                }
+            })
+            .collect::<HashSet<_>>();
+
+        // Ignore dates are evaluated after normalizing candidate timestamps to
+        // the configured service timezone, so administrators should provide
+        // local calendar dates rather than UTC midnights.
+        let ignore_dates = config
+            .issued_at_ignore_dates
+            .iter()
+            .filter_map(|value| {
+                let trimmed = value.trim();
+                if trimmed.is_empty() {
+                    return None;
+                }
+                match NaiveDate::parse_from_str(trimmed, "%Y-%m-%d") {
+                    Ok(date) => Some(date),
+                    Err(err) => {
+                        warn!(value = trimmed, error = %err, "invalid issued_at ignore date");
+                        None
+                    }
+                }
+            })
+            .collect();
+
+        Self {
+            timezone,
+            date_order,
+            filename_date_order,
+            locales,
+            ignore_dates,
+            min_date: *MIN_ISSUED_AT_DATE,
+        }
+    }
+
+    /// Returns the immutable (lowercase) locale allowlist supplied via config.
+    pub fn locales(&self) -> &HashSet<String> {
+        &self.locales
+    }
+
+    /// Returns the immutable set of local-calendar dates that should be ignored.
+    pub fn ignore_dates(&self) -> &HashSet<NaiveDate> {
+        &self.ignore_dates
+    }
+
+    /// Returns the configured service timezone (copy type).
+    pub fn timezone(&self) -> Tz {
+        self.timezone
+    }
+
+    pub fn date_order(&self) -> DateOrder {
+        self.date_order
+    }
+
+    pub fn filename_date_order(&self) -> Option<DateOrder> {
+        self.filename_date_order
+    }
+
+    pub fn normalize_naive(
+        &self,
+        date: NaiveDate,
+        now_utc: chrono::DateTime<Utc>,
+    ) -> Option<DateTime<Utc>> {
+        if !self.is_valid_with_now(date, now_utc) {
+            return None;
+        }
+        self.timezone
+            .with_ymd_and_hms(date.year(), date.month(), date.day(), 0, 0, 0)
+            .earliest()
+            .map(|dt| dt.with_timezone(&Utc))
+    }
+
+    pub fn normalize_datetime(
+        &self,
+        dt: chrono::DateTime<Utc>,
+        now_utc: chrono::DateTime<Utc>,
+    ) -> Option<DateTime<Utc>> {
+        let local_date = dt.with_timezone(&self.timezone).date_naive();
+        self.normalize_naive(local_date, now_utc)
+    }
+
+    fn is_valid_with_now(&self, date: NaiveDate, now_utc: chrono::DateTime<Utc>) -> bool {
+        if date < self.min_date {
+            return false;
+        }
+        let now_local = now_utc.with_timezone(&self.timezone).date_naive();
+        if date > now_local {
+            return false;
+        }
+        !self.ignore_dates.contains(&date)
+    }
+}
diff --git a/backend/src/lib.rs b/backend/src/lib.rs
index 10ea765..5162182 100644
--- a/backend/src/lib.rs
+++ b/backend/src/lib.rs
@@ -4,6 +4,7 @@ pub mod db;
 pub mod documents;
 pub mod error;
 pub mod http;
+pub mod issued_at;
 pub mod jobs;
 pub mod models;
 pub mod openapi;
diff --git a/backend/src/state.rs b/backend/src/state.rs
index 2526d74..484fcbe 100644
--- a/backend/src/state.rs
+++ b/backend/src/state.rs
@@ -11,6 +11,7 @@ use crate::{
     config::AppConfig,
     db::PgPool,
     error::{AppError, AppResult},
+    issued_at::IssuedAtSettings,
     storage::{ObjectStorage, TenantStorage},
     tenants::{apply_tenant_guc, clear_tenant_context, clear_user_guc, TenantService},
 };
@@ -25,6 +26,7 @@ pub struct AppState {
     pub jwt: JwtService,
     pub tenants: TenantService,
     pub passkeys: Option<PasskeyService>,
+    issued_at: Arc<IssuedAtSettings>,
 }
 
 impl AppState {
@@ -47,6 +49,7 @@ impl AppState {
         storage: Arc<dyn ObjectStorage>,
         jwt: JwtService,
     ) -> Self {
+        let issued_at = Arc::new(IssuedAtSettings::from_config(&config));
         let config = Arc::new(config);
         let tenants = TenantService::new(pool.clone());
 
@@ -65,6 +68,7 @@ impl AppState {
             jwt,
             tenants,
             passkeys,
+            issued_at,
         }
     }
 
@@ -96,4 +100,8 @@ impl AppState {
             AppError::internal("tenant storage error")
         })
     }
+
+    pub fn issued_at_settings(&self) -> Arc<IssuedAtSettings> {
+        self.issued_at.clone()
+    }
 }
diff --git a/backend/src/test_support/mod.rs b/backend/src/test_support/mod.rs
index 8b60d3e..a0ce6e6 100644
--- a/backend/src/test_support/mod.rs
+++ b/backend/src/test_support/mod.rs
@@ -176,6 +176,11 @@ impl TestApp {
             quickwit_index: None,
             worker_max_document_bytes: 200 * 1024 * 1024,
             upload_body_limit_bytes: 128 * 1024 * 1024,
+            service_timezone: "UTC".to_string(),
+            issued_at_date_order: "DMY".to_string(),
+            issued_at_filename_date_order: None,
+            issued_at_date_parser_locales: Vec::new(),
+            issued_at_ignore_dates: Vec::new(),
             webauthn_rp_id: Some("localhost".to_string()),
             webauthn_origin: Some("http://localhost".to_string()),
             webauthn_rp_name: "Papercrate".to_string(),
diff --git a/backend/src/workers/analyze.rs b/backend/src/workers/analyze.rs
index d997c9a..344cd87 100644
--- a/backend/src/workers/analyze.rs
+++ b/backend/src/workers/analyze.rs
@@ -11,6 +11,7 @@ use crate::{
 
 use super::{
     index::IndexDocumentTask,
+    issued_at::DetermineIssuedAtTask,
     job_execution_from_task_error,
     ocr::{GenerateOcrTask, OCR_TEXT_ASSET_TYPE},
     taskflow::{
@@ -115,6 +116,8 @@ impl TaskPlanner<DocumentVersionTaskContext> for AnalyzePlanner {
             }
         }
 
+        tasks.push(Box::new(DetermineIssuedAtTask::new()));
+
         if should_index {
             tasks.push(Box::new(IndexDocumentTask::new()));
         }
diff --git a/backend/src/workers/issued_at.rs b/backend/src/workers/issued_at.rs
new file mode 100644
index 0000000..216597f
--- /dev/null
+++ b/backend/src/workers/issued_at.rs
@@ -0,0 +1,697 @@
+use std::time::Duration;
+
+use async_trait::async_trait;
+use chrono::{DateTime, Datelike, NaiveDate, NaiveDateTime, TimeZone, Utc};
+use diesel::prelude::*;
+use once_cell::sync::Lazy;
+use regex::Match;
+use regex::Regex;
+use serde_json::Value;
+use tracing::{info, warn};
+use uuid::Uuid;
+
+use crate::issued_at::{DateOrder, IssuedAtSettings};
+use crate::schema::documents::dsl as documents_dsl;
+use crate::workers::ocr::OCR_TEXT_ASSET_TYPE;
+use crate::workers::taskflow::document::DocumentVersionTaskContext;
+use crate::workers::taskflow::{Task, TaskContext, TaskError, TaskResult};
+
+#[path = "issued_at_months.rs"]
+mod issued_at_months;
+
+const MAX_FILENAME_CHARS: usize = 256;
+const MAX_TEXT_CHARS: usize = 50_000;
+const DATE_SEP_PATTERN: &str = r"[\s._/\-]+";
+
+static YMD_RE: Lazy<Regex> =
+    Lazy::new(|| Regex::new(r"(?u)\b(\d{4})[./-](\d{1,2})[./-](\d{1,2})\b").expect("ymd regex"));
+
+static NUMERIC_RE: Lazy<Regex> = Lazy::new(|| {
+    Regex::new(r"(?u)\b(\d{1,2})[./-](\d{1,2})[./-](\d{2,4})\b").expect("numeric regex")
+});
+
+static DAY_MONTH_RE: Lazy<Regex> = Lazy::new(|| {
+    Regex::new(&format!(
+        r"(?u)\b(\d{{1,2}})(?:st|nd|rd|th)?{SEP}({MONTH_PATTERN}){SEP}(\d{{2,4}})\b",
+        SEP = DATE_SEP_PATTERN,
+        MONTH_PATTERN = month_pattern()
+    ))
+    .expect("day month regex")
+});
+
+static MONTH_DAY_RE: Lazy<Regex> = Lazy::new(|| {
+    Regex::new(&format!(
+        r"(?u)\b({MONTH_PATTERN}){SEP}(\d{{1,2}})(?:st|nd|rd|th)?(?:,)?{SEP}(\d{{2,4}})\b",
+        SEP = DATE_SEP_PATTERN,
+        MONTH_PATTERN = month_pattern()
+    ))
+    .expect("month day regex")
+});
+
+static MONTH_YEAR_RE: Lazy<Regex> = Lazy::new(|| {
+    Regex::new(&format!(
+        r"(?u)\b({MONTH_PATTERN})[\s._-]*(\d{{4}})\b",
+        MONTH_PATTERN = month_pattern()
+    ))
+    .expect("month year regex")
+});
+
+fn month_pattern() -> &'static str {
+    issued_at_months::pattern()
+}
+
+pub struct DetermineIssuedAtTask;
+
+impl DetermineIssuedAtTask {
+    pub fn new() -> Self {
+        Self
+    }
+}
+
+#[async_trait]
+impl Task<DocumentVersionTaskContext> for DetermineIssuedAtTask {
+    fn name(&self) -> &'static str {
+        "determine-issued-at"
+    }
+
+    async fn execute(&self, ctx: &mut DocumentVersionTaskContext) -> TaskResult<()> {
+        let document = ctx.document().await?.clone();
+        if document.issued_at.is_some() {
+            return Ok(());
+        }
+
+        let version = ctx.version().await?.clone();
+        let settings = ctx.state().issued_at_settings();
+        let now_utc = Utc::now();
+
+        let parser_hint = parser_supplied_date(&document.metadata, &version.metadata, &settings)
+            .and_then(|dt| settings.normalize_datetime(dt, now_utc));
+
+        let filename_candidate = settings.filename_date_order().and_then(|order| {
+            let normalized = normalize_content(&document.original_name, MAX_FILENAME_CHARS);
+            find_date_in_text(&normalized, order, &settings, now_utc)
+        });
+
+        let text_candidate = load_document_text(ctx).await?.and_then(|text| {
+            let normalized = normalize_content(&text, MAX_TEXT_CHARS);
+            find_date_in_text(&normalized, settings.date_order(), &settings, now_utc)
+        });
+
+        if let Some((final_date, source)) = parser_hint
+            .map(|dt| (dt, IssuedAtSource::Parser))
+            .or_else(|| filename_candidate.map(|dt| (dt, IssuedAtSource::Filename)))
+            .or_else(|| text_candidate.map(|dt| (dt, IssuedAtSource::Text)))
+        {
+            persist_issued_at(ctx, document.id, final_date.naive_utc()).await?;
+            info!(
+                job_id = %ctx.job_id(),
+                document_id = %document.id,
+                issued_at = %final_date,
+                source = source.as_ref(),
+                "issued_at determined"
+            );
+        } else {
+            info!(
+                job_id = %ctx.job_id(),
+                document_id = %document.id,
+                "no issued_at signals discovered; leaving unset"
+            );
+        }
+
+        Ok(())
+    }
+}
+
+async fn persist_issued_at(
+    ctx: &DocumentVersionTaskContext,
+    document_id: Uuid,
+    issued_at: NaiveDateTime,
+) -> TaskResult<()> {
+    let tenant_id = ctx.tenant_id();
+    let state = ctx.state().clone();
+    let result = tokio::task::spawn_blocking(move || -> Result<(), String> {
+        let mut conn = state
+            .db_for_tenant(tenant_id)
+            .map_err(|err| format!("failed to scope connection: {err:?}"))?;
+        diesel::update(
+            documents_dsl::documents
+                .filter(documents_dsl::tenant_id.eq(tenant_id))
+                .filter(documents_dsl::id.eq(document_id)),
+        )
+        .set(documents_dsl::issued_at.eq(issued_at))
+        .execute(&mut conn)
+        .map_err(|err| format!("failed to update issued_at: {err}"))?;
+        Ok(())
+    })
+    .await
+    .map_err(|err| {
+        TaskError::retry(
+            Duration::from_secs(30),
+            format!("issued_at task panicked: {err}"),
+        )
+    })?;
+    result.map_err(|err| TaskError::retry(Duration::from_secs(30), err))?;
+    Ok(())
+}
+
+fn parser_supplied_date(
+    document_meta: &Value,
+    version_meta: &Value,
+    settings: &IssuedAtSettings,
+) -> Option<DateTime<Utc>> {
+    metadata_datetime(document_meta)
+        .or_else(|| metadata_datetime(version_meta))
+        .and_then(|raw| parse_hint_datetime(raw, settings))
+}
+
+fn metadata_datetime(value: &Value) -> Option<&str> {
+    match value {
+        Value::String(s) => Some(s.as_str()),
+        Value::Object(map) => {
+            for key in [
+                "issued_at_override",
+                "issued_at",
+                "source_date",
+                "created_at",
+            ] {
+                if let Some(Value::String(s)) = map.get(key) {
+                    return Some(s.as_str());
+                }
+            }
+            if let Some(Value::Object(parser)) = map.get("parser") {
+                if let Some(Value::String(s)) = parser.get("issued_at") {
+                    return Some(s.as_str());
+                }
+            }
+            None
+        }
+        _ => None,
+    }
+}
+
+fn parse_hint_datetime(raw: &str, settings: &IssuedAtSettings) -> Option<DateTime<Utc>> {
+    if let Ok(dt) = DateTime::parse_from_rfc3339(raw) {
+        return Some(dt.with_timezone(&Utc));
+    }
+
+    if let Ok(date) = NaiveDate::parse_from_str(raw, "%Y-%m-%d") {
+        return settings
+            .timezone()
+            .with_ymd_and_hms(date.year(), date.month(), date.day(), 0, 0, 0)
+            .single()
+            .map(|dt| dt.with_timezone(&Utc));
+    }
+
+    None
+}
+
+fn normalize_content(input: &str, limit: usize) -> String {
+    let truncated: String = input.chars().take(limit).collect();
+    let mut normalized = String::with_capacity(truncated.len());
+    for ch in truncated.chars() {
+        if ch.is_control() {
+            normalized.push(' ');
+        } else {
+            normalized.extend(ch.to_lowercase());
+        }
+    }
+    normalized
+}
+
+fn find_date_in_text(
+    text: &str,
+    order: DateOrder,
+    settings: &IssuedAtSettings,
+    now_utc: DateTime<Utc>,
+) -> Option<DateTime<Utc>> {
+    collect_matches_with_spans(text, order, settings, now_utc)
+        .into_iter()
+        .map(|(_, dt)| dt)
+        .next()
+}
+
+fn collect_matches_with_spans(
+    text: &str,
+    order: DateOrder,
+    settings: &IssuedAtSettings,
+    now_utc: DateTime<Utc>,
+) -> Vec<(usize, DateTime<Utc>)> {
+    let mut matches: Vec<(usize, usize, DateTime<Utc>)> = Vec::new();
+    let mut push_date = |span: Option<Match>, date: NaiveDate| {
+        if let Some(dt) = settings.normalize_naive(date, now_utc) {
+            if let Some(span) = span {
+                let start = span.start();
+                let end = span.end();
+                if let Some(existing) =
+                    matches
+                        .iter_mut()
+                        .find(|(existing_start, existing_end, _)| {
+                            *existing_start != usize::MAX
+                                && start < *existing_end
+                                && *existing_start < end
+                        })
+                {
+                    let existing_len = existing.1.saturating_sub(existing.0);
+                    let new_len = end.saturating_sub(start);
+                    if new_len > existing_len {
+                        *existing = (start, end, dt);
+                    }
+                    return;
+                }
+                matches.push((start, end, dt));
+            } else {
+                matches.push((usize::MAX, usize::MAX, dt));
+            }
+        }
+    };
+
+    for caps in YMD_RE.captures_iter(text) {
+        let (Some(year_match), Some(month_match), Some(day_match)) =
+            (caps.get(1), caps.get(2), caps.get(3))
+        else {
+            continue;
+        };
+        let year = match year_match.as_str().parse::<i32>() {
+            Ok(value) => value,
+            Err(_) => continue,
+        };
+        let month = match month_match.as_str().parse::<u32>() {
+            Ok(value) => value,
+            Err(_) => continue,
+        };
+        let day = match day_match.as_str().parse::<u32>() {
+            Ok(value) => value,
+            Err(_) => continue,
+        };
+        if let Some(date) = NaiveDate::from_ymd_opt(year, month, day) {
+            push_date(caps.get(0), date);
+        }
+    }
+
+    for caps in NUMERIC_RE.captures_iter(text) {
+        let (Some(first_match), Some(second_match), Some(year_match)) =
+            (caps.get(1), caps.get(2), caps.get(3))
+        else {
+            continue;
+        };
+        let first = match first_match.as_str().parse::<u32>() {
+            Ok(value) => value,
+            Err(_) => continue,
+        };
+        let second = match second_match.as_str().parse::<u32>() {
+            Ok(value) => value,
+            Err(_) => continue,
+        };
+        let year_raw = year_match.as_str();
+        let mut year = match year_raw.parse::<i32>() {
+            Ok(value) => value,
+            Err(_) => continue,
+        };
+        if year_raw.len() == 2 {
+            year += if year >= 70 { 1900 } else { 2000 };
+        }
+        // NUMERIC_RE always captures a day-first form (dd[sep]mm[sep]yy(yy));
+        // YMD layouts are handled earlier by YMD_RE, so YMD here is treated the
+        // same as DMY to avoid mis-parsing strings like 01-07-2024.
+        let (day, month) = match order {
+            DateOrder::Dmy | DateOrder::Ymd => (first, second),
+            DateOrder::Mdy => (second, first),
+        };
+        if let Some(date) = NaiveDate::from_ymd_opt(year, month, day) {
+            push_date(caps.get(0), date);
+        }
+    }
+
+    for caps in DAY_MONTH_RE.captures_iter(text) {
+        let (Some(day_match), Some(month_match), Some(year_match)) =
+            (caps.get(1), caps.get(2), caps.get(3))
+        else {
+            continue;
+        };
+        let day_str = day_match
+            .as_str()
+            .trim_matches(|c: char| !c.is_ascii_digit());
+        let day = match day_str.parse::<u32>() {
+            Ok(value) => value,
+            Err(_) => continue,
+        };
+        let Some(month) = month_name_to_number(month_match.as_str(), settings) else {
+            continue;
+        };
+        if year_match.as_str().len() < 3 {
+            continue;
+        }
+        let Some(year) = normalize_year(year_match.as_str()) else {
+            continue;
+        };
+        if let Some(date) = NaiveDate::from_ymd_opt(year, month, day) {
+            push_date(caps.get(0), date);
+        }
+    }
+
+    for caps in MONTH_DAY_RE.captures_iter(text) {
+        let (Some(month_match), Some(day_match), Some(year_match)) =
+            (caps.get(1), caps.get(2), caps.get(3))
+        else {
+            continue;
+        };
+        let Some(month) = month_name_to_number(month_match.as_str(), settings) else {
+            continue;
+        };
+        let day_str = day_match
+            .as_str()
+            .trim_matches(|c: char| !c.is_ascii_digit());
+        let day = match day_str.parse::<u32>() {
+            Ok(value) => value,
+            Err(_) => continue,
+        };
+        let Some(year) = normalize_year(year_match.as_str()) else {
+            continue;
+        };
+        if let Some(date) = NaiveDate::from_ymd_opt(year, month, day) {
+            push_date(caps.get(0), date);
+        }
+    }
+
+    for caps in MONTH_YEAR_RE.captures_iter(text) {
+        let (Some(month_match), Some(year_match)) = (caps.get(1), caps.get(2)) else {
+            continue;
+        };
+        let Some(month) = month_name_to_number(month_match.as_str(), settings) else {
+            continue;
+        };
+        let year = match year_match.as_str().parse::<i32>() {
+            Ok(value) => value,
+            Err(_) => continue,
+        };
+        if let Some(date) = NaiveDate::from_ymd_opt(year, month, 1) {
+            push_date(caps.get(0), date);
+        }
+    }
+
+    matches.sort_by_key(|(start, _, _)| *start);
+    matches
+        .into_iter()
+        .map(|(start, _, dt)| (start, dt))
+        .collect()
+}
+
+fn normalize_year(raw: &str) -> Option<i32> {
+    if raw.len() == 2 {
+        let mut year = raw.parse::<i32>().ok()?;
+        year += if year >= 70 { 1900 } else { 2000 };
+        Some(year)
+    } else {
+        raw.parse::<i32>().ok()
+    }
+}
+
+fn month_name_to_number(value: &str, settings: &IssuedAtSettings) -> Option<u32> {
+    let normalized = value.trim();
+    let variant = issued_at_months::variants()
+        .iter()
+        .find(|entry| entry.name == normalized)?;
+    if settings.locales().is_empty() || variant.locales.is_empty() {
+        return Some(variant.month);
+    }
+    if variant
+        .locales
+        .iter()
+        .any(|locale| settings.locales().contains(*locale))
+    {
+        Some(variant.month)
+    } else {
+        None
+    }
+}
+
+async fn load_document_text(ctx: &mut DocumentVersionTaskContext) -> TaskResult<Option<String>> {
+    let object_key = {
+        let assets = ctx.assets().await?;
+        assets
+            .get(OCR_TEXT_ASSET_TYPE)
+            .and_then(|asset| asset.objects.first())
+            .map(|object| object.s3_key.clone())
+    };
+
+    let Some(key) = object_key else {
+        return Ok(None);
+    };
+
+    match ctx.storage().get_object(&key).await {
+        Ok(bytes) => match String::from_utf8(bytes) {
+            Ok(mut text) => {
+                if text.len() > MAX_TEXT_CHARS {
+                    text.truncate(MAX_TEXT_CHARS);
+                }
+                Ok(Some(text))
+            }
+            Err(err) => {
+                warn!(error = %err, "ocr text asset not valid utf-8");
+                Ok(None)
+            }
+        },
+        Err(err) => {
+            warn!(error = %err, "failed to download ocr text for issued_at extractor");
+            Ok(None)
+        }
+    }
+}
+
+#[derive(Copy, Clone)]
+enum IssuedAtSource {
+    Parser,
+    Filename,
+    Text,
+}
+
+impl IssuedAtSource {
+    fn as_ref(&self) -> &'static str {
+        match self {
+            IssuedAtSource::Parser => "parser",
+            IssuedAtSource::Filename => "filename",
+            IssuedAtSource::Text => "text",
+        }
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use crate::config::AppConfig;
+    use serde::Deserialize;
+    use serde_yaml::Value as YamlValue;
+
+    use once_cell::sync::Lazy;
+
+    #[derive(Clone, Deserialize)]
+    struct CaseSuite {
+        cases: Vec<CaseDefinition>,
+    }
+
+    #[derive(Clone, Deserialize)]
+    struct CaseDefinition {
+        name: String,
+        parser: String,
+        #[serde(default)]
+        filename: Option<String>,
+        #[serde(default)]
+        content: Option<String>,
+        #[serde(default)]
+        settings: CaseSettings,
+        expected: ExpectedCase,
+    }
+
+    #[derive(Clone, Default, Deserialize)]
+    struct CaseSettings {
+        #[serde(rename = "DATE_PARSER_LANGUAGES", default)]
+        date_parser_languages: Vec<String>,
+        #[serde(rename = "FILENAME_DATE_ORDER")]
+        filename_date_order: Option<String>,
+        #[serde(rename = "DATE_ORDER")]
+        date_order: Option<String>,
+        #[serde(rename = "IGNORE_DATES", default)]
+        ignore_dates: Vec<String>,
+    }
+
+    #[derive(Clone, Deserialize)]
+    struct ExpectedCase {
+        mode: ExpectedMode,
+        #[serde(default)]
+        value: Option<YamlValue>,
+    }
+
+    #[derive(Clone, Deserialize, PartialEq)]
+    #[serde(rename_all = "lowercase")]
+    enum ExpectedMode {
+        None,
+        Single,
+        Multiple,
+    }
+
+    static CASES: Lazy<CaseSuite> = Lazy::new(|| {
+        let raw = include_str!("../../tests/data/issued_at_cases.yaml");
+        serde_yaml::from_str(raw).expect("failed to parse issued_at cases")
+    });
+
+    pub(crate) fn run_named_case(name: &str) {
+        let case = CASES
+            .cases
+            .iter()
+            .find(|case| case.name == name)
+            .unwrap_or_else(|| panic!("case '{}' not found", name))
+            .clone();
+        run_case(case);
+    }
+
+    fn run_case(case: CaseDefinition) {
+        let mut config = base_config();
+        if let Some(order) = case.settings.date_order {
+            config.issued_at_date_order = order;
+        }
+        config.issued_at_filename_date_order = case.settings.filename_date_order;
+        config.issued_at_date_parser_locales = case.settings.date_parser_languages;
+        config.issued_at_ignore_dates = case.settings.ignore_dates;
+
+        let settings = IssuedAtSettings::from_config(&config);
+        let now_utc = Utc.with_ymd_and_hms(2025, 1, 1, 0, 0, 0).single().unwrap();
+        let mut matches = Vec::new();
+
+        if let Some(filename) = &case.filename {
+            if let Some(order) = settings.filename_date_order() {
+                let normalized = normalize_content(filename, MAX_FILENAME_CHARS);
+                matches.extend(
+                    collect_matches_with_spans(&normalized, order, &settings, now_utc)
+                        .into_iter()
+                        .map(|(_, dt)| dt.date_naive()),
+                );
+            }
+        }
+
+        if let Some(content) = &case.content {
+            let normalized = normalize_content(content, MAX_TEXT_CHARS);
+            matches.extend(
+                collect_matches_with_spans(&normalized, settings.date_order(), &settings, now_utc)
+                    .into_iter()
+                    .map(|(_, dt)| dt.date_naive()),
+            );
+        }
+
+        let actual: Vec<String> = matches
+            .into_iter()
+            .map(|date| date.format("%Y-%m-%d").to_string())
+            .collect();
+
+        match case.parser.as_str() {
+            "parse_date" => match case.expected.mode {
+                ExpectedMode::None => assert!(
+                    actual.is_empty(),
+                    "case '{}' expected no matches, got {:?}",
+                    case.name,
+                    actual
+                ),
+                ExpectedMode::Single => {
+                    let expected = case.expected.single_value();
+                    assert!(
+                        expected.is_some(),
+                        "case '{}' is missing expected single value",
+                        case.name
+                    );
+                    assert_eq!(
+                        actual.first(),
+                        expected.as_ref(),
+                        "case '{}' single mismatch",
+                        case.name
+                    );
+                }
+                ExpectedMode::Multiple => panic!(
+                    "case '{}' declares parse_date but expects multiple results",
+                    case.name
+                ),
+            },
+            "parse_date_generator" => {
+                let expected = case.expected.multiple_values().unwrap_or_default();
+                assert_eq!(expected, actual, "case '{}' multiple mismatch", case.name);
+            }
+            other => panic!("unsupported parser '{}' in case {}", other, case.name),
+        }
+    }
+
+    impl ExpectedCase {
+        fn single_value(&self) -> Option<String> {
+            match self.value.as_ref()? {
+                YamlValue::String(value) => Some(value.clone()),
+                other => Some(other.as_str()?.to_string()),
+            }
+        }
+
+        fn multiple_values(&self) -> Option<Vec<String>> {
+            let list = match self.value.as_ref()? {
+                YamlValue::Sequence(seq) => seq,
+                _ => return None,
+            };
+            Some(
+                list.iter()
+                    .filter_map(|value| value.as_str().map(|s| s.to_string()))
+                    .collect(),
+            )
+        }
+    }
+
+    fn base_config() -> AppConfig {
+        AppConfig {
+            database_url: "postgres://test".to_string(),
+            database_max_pool_size: 5,
+            server_host: "127.0.0.1".to_string(),
+            server_port: 0,
+            webdav_host: "127.0.0.1".to_string(),
+            webdav_port: 0,
+            jwt_secret: "secret".to_string(),
+            jwt_issuer: "issuer".to_string(),
+            jwt_audience: "audience".to_string(),
+            jwt_expiry_minutes: 60,
+            download_token_audience: "download".to_string(),
+            download_token_expiry_minutes: 60,
+            refresh_token_expiry_days: 30,
+            refresh_cookie_secure: false,
+            refresh_cookie_domain: None,
+            cors_allowed_origin: None,
+            proxy_downloads: false,
+            aws_endpoint_url: None,
+            aws_access_key_id: None,
+            aws_secret_access_key: None,
+            aws_region: "us-east-1".to_string(),
+            s3_bucket: "bucket".to_string(),
+            quickwit_endpoint: None,
+            quickwit_index: None,
+            worker_max_document_bytes: 100 * 1024 * 1024,
+            upload_body_limit_bytes: 64 * 1024 * 1024,
+            service_timezone: "UTC".to_string(),
+            issued_at_date_order: "DMY".to_string(),
+            issued_at_filename_date_order: None,
+            issued_at_date_parser_locales: Vec::new(),
+            issued_at_ignore_dates: Vec::new(),
+            webauthn_rp_id: Some("localhost".to_string()),
+            webauthn_origin: Some("http://localhost".to_string()),
+            webauthn_rp_name: "Papercrate".to_string(),
+        }
+    }
+
+    include!(concat!(env!("OUT_DIR"), "/issued_at_generated_tests.rs"));
+
+    #[test]
+    fn month_name_lookup_handles_turkish_variants() {
+        let settings = IssuedAtSettings::from_config(&base_config());
+        assert_eq!(month_name_to_number("şubat", &settings), Some(2));
+        assert_eq!(month_name_to_number("subat", &settings), Some(2));
+    }
+
+    #[test]
+    fn locale_filter_limits_month_names() {
+        let mut config = base_config();
+        config.issued_at_date_parser_locales = vec!["tr".into()];
+        let settings = IssuedAtSettings::from_config(&config);
+        assert_eq!(month_name_to_number("january", &settings), None);
+        assert_eq!(month_name_to_number("şubat", &settings), Some(2));
+    }
+}
diff --git a/backend/src/workers/issued_at_months.rs b/backend/src/workers/issued_at_months.rs
new file mode 100644
index 0000000..dc648a3
--- /dev/null
+++ b/backend/src/workers/issued_at_months.rs
@@ -0,0 +1,15 @@
+pub(super) struct MonthVariant {
+    pub name: &'static str,
+    pub month: u32,
+    pub locales: &'static [&'static str],
+}
+
+include!(concat!(env!("OUT_DIR"), "/issued_at_months.rs"));
+
+pub(super) fn pattern() -> &'static str {
+    MONTH_PATTERN
+}
+
+pub(super) fn variants() -> &'static [MonthVariant] {
+    MONTH_VARIANTS
+}
diff --git a/backend/src/workers/mod.rs b/backend/src/workers/mod.rs
index 333046e..650667c 100644
--- a/backend/src/workers/mod.rs
+++ b/backend/src/workers/mod.rs
@@ -16,6 +16,7 @@ use taskflow::TaskError;
 pub mod analyze;
 pub mod common;
 pub mod index;
+pub mod issued_at;
 pub mod ocr;
 pub mod purge;
 pub mod taskflow;
diff --git a/backend/tests/data/issued_at_cases.yaml b/backend/tests/data/issued_at_cases.yaml
new file mode 100644
index 0000000..8e58004
--- /dev/null
+++ b/backend/tests/data/issued_at_cases.yaml
@@ -0,0 +1,582 @@
+cases:
+  - name: date_format_1
+    parser: parse_date
+    filename: null
+    content: "lorem ipsum 130218 lorem ipsum"
+    settings: {}
+    expected:
+      mode: none
+
+  - name: date_format_2
+    parser: parse_date
+    filename: null
+    content: "lorem ipsum 2018 lorem ipsum"
+    settings: {}
+    expected:
+      mode: none
+
+  - name: date_format_3
+    parser: parse_date
+    filename: null
+    content: "lorem ipsum 20180213 lorem ipsum"
+    settings: {}
+    expected:
+      mode: none
+
+  - name: date_format_4
+    parser: parse_date
+    filename: null
+    content: "lorem ipsum 13.02.2018 lorem ipsum"
+    settings: {}
+    expected:
+      mode: single
+      value: 2018-02-13
+
+  - name: date_format_5
+    parser: parse_date
+    filename: null
+    content: "lorem ipsum 130218, 2018, 20180213 and lorem 13.02.2018 lorem ipsum"
+    settings: {}
+    expected:
+      mode: single
+      value: 2018-02-13
+
+  - name: date_format_6
+    parser: parse_date
+    filename: null
+    content: |
+      lorem ipsum
+      Wohnort
+      3100
+      IBAN
+      AT87 4534
+      1234
+      1234 5678
+      BIC
+      lorem ipsum
+    settings: {}
+    expected:
+      mode: none
+
+  - name: date_format_7
+    parser: parse_date
+    filename: null
+    content: |
+      lorem ipsum
+      März 2019
+      lorem ipsum
+    settings:
+      DATE_PARSER_LANGUAGES:
+        - de
+    expected:
+      mode: single
+      value: 2019-03-01
+
+  - name: date_format_8
+    parser: parse_date
+    filename: null
+    content: |
+      lorem ipsum
+      Wohnort
+      3100
+      IBAN
+      AT87 4534
+      1234
+      1234 5678
+      BIC
+      lorem ipsum
+      März 2020
+    settings:
+      DATE_PARSER_LANGUAGES:
+        - de
+    expected:
+      mode: single
+      value: 2020-03-01
+
+  - name: date_format_9
+    parser: parse_date
+    filename: null
+    content: |
+      lorem ipsum
+      27. Nullmonth 2020
+      März 2020
+      lorem ipsum
+    settings:
+      DATE_PARSER_LANGUAGES:
+        - de
+    expected:
+      mode: single
+      value: 2020-03-01
+
+  - name: date_format_10
+    parser: parse_date
+    filename: null
+    content: "Customer Number Currency 22-MAR-2022 Credit Card 1934829304"
+    settings: {}
+    expected:
+      mode: single
+      value: 2022-03-22
+
+  - name: date_format_11
+    parser: parse_date
+    filename: null
+    content: "Customer Number Currency 22 MAR 2022 Credit Card 1934829304"
+    settings: {}
+    expected:
+      mode: single
+      value: 2022-03-22
+
+  - name: date_format_12
+    parser: parse_date
+    filename: null
+    content: "Customer Number Currency 22/MAR/2022 Credit Card 1934829304"
+    settings: {}
+    expected:
+      mode: single
+      value: 2022-03-22
+
+  - name: date_format_13
+    parser: parse_date
+    filename: null
+    content: "Customer Number Currency 22.MAR.2022 Credit Card 1934829304"
+    settings: {}
+    expected:
+      mode: single
+      value: 2022-03-22
+
+  - name: date_format_14
+    parser: parse_date
+    filename: null
+    content: "Customer Number Currency 22.MAR 2022 Credit Card 1934829304"
+    settings: {}
+    expected:
+      mode: single
+      value: 2022-03-22
+
+  - name: date_format_15
+    parser: parse_date
+    filename: null
+    content: "Customer Number Currency 22.MAR.22 Credit Card 1934829304"
+    settings: {}
+    expected:
+      mode: none
+
+  - name: date_format_16
+    parser: parse_date
+    filename: null
+    content: "Customer Number Currency 22.MAR,22 Credit Card 1934829304"
+    settings: {}
+    expected:
+      mode: none
+
+  - name: date_format_17
+    parser: parse_date
+    filename: null
+    content: "Customer Number Currency 22,MAR,2022 Credit Card 1934829304"
+    settings: {}
+    expected:
+      mode: none
+
+  - name: date_format_18
+    parser: parse_date
+    filename: null
+    content: "Customer Number Currency 22 MAR,2022 Credit Card 1934829304"
+    settings: {}
+    expected:
+      mode: none
+
+  - name: date_format_19
+    parser: parse_date
+    filename: null
+    content: "Customer Number Currency 21st MAR 2022 Credit Card 1934829304"
+    settings: {}
+    expected:
+      mode: single
+      value: 2022-03-21
+
+  - name: date_format_20
+    parser: parse_date
+    filename: null
+    content: "Customer Number Currency 22nd March 2022 Credit Card 1934829304"
+    settings: {}
+    expected:
+      mode: single
+      value: 2022-03-22
+
+  - name: date_format_21
+    parser: parse_date
+    filename: null
+    content: "Customer Number Currency 2nd MAR 2022 Credit Card 1934829304"
+    settings: {}
+    expected:
+      mode: single
+      value: 2022-03-02
+
+  - name: date_format_22
+    parser: parse_date
+    filename: null
+    content: "Customer Number Currency 23rd MAR 2022 Credit Card 1934829304"
+    settings: {}
+    expected:
+      mode: single
+      value: 2022-03-23
+
+  - name: date_format_23
+    parser: parse_date
+    filename: null
+    content: "Customer Number Currency 24th MAR 2022 Credit Card 1934829304"
+    settings: {}
+    expected:
+      mode: single
+      value: 2022-03-24
+
+  - name: date_format_24
+    parser: parse_date
+    filename: null
+    content: "Customer Number Currency 21-MAR-2022 Credit Card 1934829304"
+    settings: {}
+    expected:
+      mode: single
+      value: 2022-03-21
+
+  - name: date_format_25
+    parser: parse_date
+    filename: null
+    content: "Customer Number Currency 25TH MAR 2022 Credit Card 1934829304"
+    settings: {}
+    expected:
+      mode: single
+      value: 2022-03-25
+
+  - name: date_format_26
+    parser: parse_date
+    filename: null
+    content: "CHASE 0 September 25, 2019 JPMorgan Chase Bank, NA. P0 Box 182051"
+    settings: {}
+    expected:
+      mode: single
+      value: 2019-09-25
+
+  - name: numeric_mdy_slash
+    parser: parse_date
+    filename: null
+    content: "03/17/2008"
+    settings:
+      DATE_ORDER: MDY
+    expected:
+      mode: single
+      value: 2008-03-17
+
+  - name: crazy_date_past
+    parser: parse_date
+    filename: null
+    content: "01-07-0590 00:00:00"
+    settings: {}
+    expected:
+      mode: none
+
+  - name: crazy_date_future
+    parser: parse_date
+    filename: null
+    content: "01-07-2350 00:00:00"
+    settings: {}
+    expected:
+      mode: none
+
+  - name: crazy_date_with_spaces
+    parser: parse_date
+    filename: null
+    content: "20 408000l 2475"
+    settings: {}
+    expected:
+      mode: none
+
+  - name: utf_month_names_decembre
+    parser: parse_date
+    filename: null
+    content: "13 décembre 2023"
+    settings:
+      DATE_PARSER_LANGUAGES:
+        - fr
+        - de
+        - hr
+        - cs
+        - pl
+        - tr
+    expected:
+      mode: single
+      value: 2023-12-13
+
+  - name: utf_month_names_aout
+    parser: parse_date
+    filename: null
+    content: "13 août 2022"
+    settings:
+      DATE_PARSER_LANGUAGES:
+        - fr
+        - de
+        - hr
+        - cs
+        - pl
+        - tr
+    expected:
+      mode: single
+      value: 2022-08-13
+
+  - name: utf_month_names_marz
+    parser: parse_date
+    filename: null
+    content: "11 März 2020"
+    settings:
+      DATE_PARSER_LANGUAGES:
+        - fr
+        - de
+        - hr
+        - cs
+        - pl
+        - tr
+    expected:
+      mode: single
+      value: 2020-03-11
+
+  - name: utf_month_names_ozujka
+    parser: parse_date
+    filename: null
+    content: "17. ožujka 2018."
+    settings:
+      DATE_PARSER_LANGUAGES:
+        - fr
+        - de
+        - hr
+        - cs
+        - pl
+        - tr
+    expected:
+      mode: single
+      value: 2018-03-17
+
+  - name: utf_month_names_veljace
+    parser: parse_date
+    filename: null
+    content: "1. veljače 2016."
+    settings:
+      DATE_PARSER_LANGUAGES:
+        - fr
+        - de
+        - hr
+        - cs
+        - pl
+        - tr
+    expected:
+      mode: single
+      value: 2016-02-01
+
+  - name: utf_month_names_unora
+    parser: parse_date
+    filename: null
+    content: "15. února 1985"
+    settings:
+      DATE_PARSER_LANGUAGES:
+        - fr
+        - de
+        - hr
+        - cs
+        - pl
+        - tr
+    expected:
+      mode: single
+      value: 1985-02-15
+
+  - name: utf_month_names_zari
+    parser: parse_date
+    filename: null
+    content: "30. září 2011"
+    settings:
+      DATE_PARSER_LANGUAGES:
+        - fr
+        - de
+        - hr
+        - cs
+        - pl
+        - tr
+    expected:
+      mode: single
+      value: 2011-09-30
+
+  - name: utf_month_names_kvetna
+    parser: parse_date
+    filename: null
+    content: "28. května 1990"
+    settings:
+      DATE_PARSER_LANGUAGES:
+        - fr
+        - de
+        - hr
+        - cs
+        - pl
+        - tr
+    expected:
+      mode: single
+      value: 1990-05-28
+
+  - name: utf_month_names_grudzien
+    parser: parse_date
+    filename: null
+    content: "1. grudzień 1997"
+    settings:
+      DATE_PARSER_LANGUAGES:
+        - fr
+        - de
+        - hr
+        - cs
+        - pl
+        - tr
+    expected:
+      mode: single
+      value: 1997-12-01
+
+  - name: utf_month_names_subat
+    parser: parse_date
+    filename: null
+    content: "17 Şubat 2024"
+    settings:
+      DATE_PARSER_LANGUAGES:
+        - fr
+        - de
+        - hr
+        - cs
+        - pl
+        - tr
+    expected:
+      mode: single
+      value: 2024-02-17
+
+  - name: utf_month_names_agustos
+    parser: parse_date
+    filename: null
+    content: "30 Ağustos 2012"
+    settings:
+      DATE_PARSER_LANGUAGES:
+        - fr
+        - de
+        - hr
+        - cs
+        - pl
+        - tr
+    expected:
+      mode: single
+      value: 2012-08-30
+
+  - name: utf_month_names_eylul
+    parser: parse_date
+    filename: null
+    content: "17 Eylül 2000"
+    settings:
+      DATE_PARSER_LANGUAGES:
+        - fr
+        - de
+        - hr
+        - cs
+        - pl
+        - tr
+    expected:
+      mode: single
+      value: 2000-09-17
+
+  - name: utf_month_names_oktober
+    parser: parse_date
+    filename: null
+    content: "5. október 1992"
+    settings:
+      DATE_PARSER_LANGUAGES:
+        - fr
+        - de
+        - hr
+        - cs
+        - pl
+        - tr
+        - hu
+    expected:
+      mode: single
+      value: 1992-10-05
+
+  - name: multiple_dates
+    parser: parse_date_generator
+    filename: null
+    content: |
+      This text has multiple dates.
+      For example 02.02.2018, 22 July 2022 and December 2021.
+      But not 24-12-9999 because it's in the future...
+    settings: {}
+    expected:
+      mode: multiple
+      value:
+        - 2018-02-02
+        - 2022-07-22
+        - 2021-12-01
+
+  - name: filename_date_parse_valid_ymd
+    parser: parse_date
+    filename: /tmp/Scan-2022-04-01.pdf
+    content: "No date in here"
+    settings:
+      FILENAME_DATE_ORDER: YMD
+    expected:
+      mode: single
+      value: 2022-04-01
+
+  - name: filename_date_parse_valid_dmy
+    parser: parse_date
+    filename: /tmp/Scan-10.01.2021.pdf
+    content: "No date in here"
+    settings:
+      FILENAME_DATE_ORDER: DMY
+    expected:
+      mode: single
+      value: 2021-01-10
+
+  - name: filename_date_parse_invalid
+    parser: parse_date
+    filename: "/tmp/20 408000l 2475 - test.pdf"
+    content: "No date in here"
+    settings:
+      FILENAME_DATE_ORDER: YMD
+    expected:
+      mode: none
+
+  - name: filename_date_ignored_use_content
+    parser: parse_date
+    filename: /tmp/Scan-2022-04-01.pdf
+    content: "The matching date is 24.03.2022"
+    settings:
+      FILENAME_DATE_ORDER: YMD
+      IGNORE_DATES:
+        - 2022-04-01
+    expected:
+      mode: single
+      value: 2022-03-24
+
+  - name: ignored_dates_default_order
+    parser: parse_date
+    filename: null
+    content: "lorem ipsum 110319, 20200117 and lorem 13.02.2018 lorem ipsum"
+    settings:
+      IGNORE_DATES:
+        - 2019-11-03
+        - 2020-01-17
+    expected:
+      mode: single
+      value: 2018-02-13
+
+  - name: ignored_dates_order_ymd
+    parser: parse_date
+    filename: null
+    content: "lorem ipsum 190311, 20200117 and lorem 13.02.2018 lorem ipsum"
+    settings:
+      FILENAME_DATE_ORDER: YMD
+      IGNORE_DATES:
+        - 2019-11-03
+        - 2020-01-17
+    expected:
+      mode: single
+      value: 2018-02-13