diff --git a/tests/host_tools/test_syscalls.c b/tests/host_tools/test_syscalls.c
new file mode 100644
index 00000000000..6a58edf0983
--- /dev/null
+++ b/tests/host_tools/test_syscalls.c
@@ -0,0 +1,77 @@
+// Copyright 2024 Amazon.com, Inc. or its affiliates. All Rights Reserved.
+// SPDX-License-Identifier: Apache-2.0
+
+// This is used by `test_seccomp_validate.py`
+
+#include <linux/types.h>
+#include <linux/filter.h>
+#include <linux/seccomp.h>
+#include <sys/prctl.h>
+#include <sys/stat.h>
+
+#include <unistd.h>
+#include <stdlib.h>
+#include <stdio.h>
+#include <string.h>
+#include <fcntl.h>
+
+
+void install_bpf_filter(char *bpf_file) {
+    int fd = open(bpf_file, O_RDONLY);
+    if (fd == -1) {
+        perror("open");
+        exit(EXIT_FAILURE);
+    }
+    struct stat sb;
+    if (fstat(fd, &sb) == -1) {
+        perror("stat");
+        exit(EXIT_FAILURE);
+    }
+    size_t size = sb.st_size;
+    size_t insn_len = size / sizeof(struct sock_filter);
+    struct sock_filter *filterbuf = (struct sock_filter*)malloc(size);
+    if (read(fd, filterbuf, size) == -1) {
+        perror("read");
+        exit(EXIT_FAILURE);
+    }
+
+    /* Install seccomp filter */
+    struct sock_fprog prog = {
+        .len = (unsigned short)(insn_len),
+        .filter = filterbuf,
+    };
+    if (prctl(PR_SET_NO_NEW_PRIVS, 1, 0, 0, 0)) {
+        perror("prctl(NO_NEW_PRIVS)");
+        exit(EXIT_FAILURE);
+    }
+    if (prctl(PR_SET_SECCOMP, SECCOMP_MODE_FILTER, &prog)) {
+        perror("prctl(SECCOMP)");
+        exit(EXIT_FAILURE);
+    }
+}
+
+
+int main(int argc, char **argv) {
+    /* parse arguments */
+    if (argc < 3) {
+        fprintf(stderr, "Usage: %s BPF_FILE ARG0..\n", argv[0]);
+        exit(EXIT_FAILURE);
+    }
+    char *bpf_file = argv[1];
+    long syscall_id = atoi(argv[2]);
+    long arg0, arg1, arg2, arg3;
+    arg0 = arg1 = arg2 = arg3 = 0;
+    if (argc > 3) arg0 = atoi(argv[3]);
+    if (argc > 4) arg1 = atoi(argv[4]);
+    if (argc > 5) arg2 = atoi(argv[5]);
+    if (argc > 6) arg3 = atoi(argv[6]);
+
+    /* read seccomp filter from file */
+    if (strcmp(bpf_file, "/dev/null") != 0) {
+        install_bpf_filter(bpf_file);
+    }
+
+    long res = syscall(syscall_id, arg0, arg1, arg2, arg3);
+    printf("%ld\n", res);
+    return EXIT_SUCCESS;
+}
diff --git a/tests/integration_tests/security/test_seccomp_validate.py b/tests/integration_tests/security/test_seccomp_validate.py
new file mode 100644
index 00000000000..5a0be067899
--- /dev/null
+++ b/tests/integration_tests/security/test_seccomp_validate.py
@@ -0,0 +1,138 @@
+# Copyright 2024 Amazon.com, Inc. or its affiliates. All Rights Reserved.
+# SPDX-License-Identifier: Apache-2.0
+
+"""Test that validates that seccompiler filters work as expected"""
+
+import json
+import platform
+import resource
+import struct
+from pathlib import Path
+
+import pytest
+import seccomp
+
+from framework import utils
+from host_tools import cargo_build
+
+ARCH = platform.machine()
+
+
+@pytest.fixture(scope="session")
+def bin_test_syscall(test_fc_session_root_path):
+    """Build the test_syscall binary."""
+    test_syscall_bin = Path(test_fc_session_root_path) / "test_syscall"
+    cargo_build.gcc_compile("host_tools/test_syscalls.c", test_syscall_bin)
+    assert test_syscall_bin.exists()
+    yield test_syscall_bin
+
+
+class BpfMapReader:
+    """
+    Simple reader for the files that seccompiler-bin produces
+
+    The files are serialized with bincode[1] in format that is easy to parse.
+
+        sock_filter = <ushort uchar uchar uint>
+        BpfProgram = Vec<sock_filter>
+        BpfMap = BTreeMap(str, BpfProgram)
+        str = Vec<uchar>
+
+    [1] https://github.com/bincode-org/bincode/blob/trunk/docs/spec.md
+    """
+
+    INSN_FMT = "<HBBI"
+    INSN_SIZEOF = struct.calcsize(INSN_FMT)
+
+    def __init__(self, buf):
+        self.buf = buf
+        self.offset = 0
+
+    @classmethod
+    def from_file(cls, file):
+        """Initialize a buffer from a file"""
+        return cls(Path(file).read_bytes())
+
+    def read_format(self, fmt):
+        """Read a struct format string from the buffer"""
+        val = struct.unpack_from(fmt, self.buf, offset=self.offset)
+        self.offset += struct.calcsize(fmt)
+        if len(val) == 1:
+            return val[0]
+        return val
+
+    def is_eof(self):
+        """Are we at the end of the buffer?"""
+        return self.offset == len(self.buf)
+
+    def lookahead(self, size):
+        """Look ahead `size` bytes"""
+        return self.buf[self.offset : self.offset + size]
+
+    def split(self):
+        """Return separate filters"""
+        threads = {}
+        # how many entries in the map
+        map_len = self.read_format("<Q")
+        for _ in range(map_len):
+            # read key
+            key_str_len = self.read_format("<Q")
+            key_str = self.read_format(f"{key_str_len}s")
+            # read value: vec of instructions
+            insn_len = self.read_format("<Q")
+            data = self.lookahead(insn_len * self.INSN_SIZEOF)
+            threads[key_str.decode("ascii")] = data
+            self.offset += len(data)
+
+        assert self.is_eof()
+        return threads
+
+
+def test_validate_filter(seccompiler, bin_test_syscall, monkeypatch, tmp_path):
+    """Assert that the seccomp filter matches the JSON description."""
+
+    fc_filter_path = Path(f"../resources/seccomp/{ARCH}-unknown-linux-musl.json")
+    fc_filter = json.loads(fc_filter_path.read_text(encoding="ascii"))
+
+    # cd to a tmp dir because we may generate a bunch of intermediate files
+    monkeypatch.chdir(tmp_path)
+    # prevent coredumps
+    resource.setrlimit(resource.RLIMIT_CORE, (0, 0))
+
+    bpf_path = seccompiler.compile(fc_filter)
+    filters = BpfMapReader.from_file(bpf_path).split()
+    arch = seccomp.Arch.X86_64 if ARCH == "x86_64" else seccomp.Arch.AARCH64
+    for thread, filter_data in fc_filter.items():
+        filter_path = Path(f"{thread}.bpf")
+        filter_path.write_bytes(filters[thread])
+        # for each rule, run the helper program and execute a syscall
+        for rule in filter_data["filter"]:
+            print(filter_path, rule)
+            syscall = rule["syscall"]
+            # this one cannot be called directly
+            if syscall in ["rt_sigreturn"]:
+                continue
+            syscall_id = seccomp.resolve_syscall(arch, syscall)
+            cmd = f"{bin_test_syscall} {filter_path} {syscall_id}"
+            if "args" not in rule:
+                # syscall should be allowed with any arguments and exit 0
+                assert utils.run_cmd(cmd).returncode == 0
+            else:
+                allowed_args = [0] * 4
+                # if we call it with allowed args, it should exit 0
+                for arg in rule["args"]:
+                    allowed_args[arg["index"]] = arg["val"]
+                allowed_str = " ".join(str(x) for x in allowed_args)
+                assert utils.run_cmd(f"{cmd} {allowed_str}").returncode == 0
+                # for each allowed arg try a different number
+                for arg in rule["args"]:
+                    # We just add 1000000 to the allowed arg and assume it is
+                    # not something we allow in another rule. While not perfect
+                    # it works in practice.
+                    bad_args = allowed_args.copy()
+                    bad_args[arg["index"]] = str(arg["val"] + 1_000_000)
+                    unallowed_str = " ".join(str(x) for x in bad_args)
+                    outcome = utils.run_cmd(f"{cmd} {unallowed_str}")
+                    # if we call it with unallowed args, it should exit 159
+                    # 159 = 128 (abnormal termination) + 31 (SIGSYS)
+                    assert outcome.returncode == 159