diff --git a/mrbgems/mruby-encoding/mrbgem.rake b/mrbgems/mruby-encoding/mrbgem.rake
new file mode 100644
index 000000000..8d5e41284
--- /dev/null
+++ b/mrbgems/mruby-encoding/mrbgem.rake
@@ -0,0 +1,8 @@
+MRuby::Gem::Specification.new('mruby-encoding') do |spec|
+  spec.license = 'MIT'
+  spec.author  = 'mruby developers'
+  spec.summary = "Poorman's Encoding for mruby"
+  spec.build.defines << "HAVE_MRUBY_ENCODING_GEM"
+  spec.build.defines << "MRB_UTF8_STRING"
+  spec.add_test_dependency 'mruby-string-ext'
+end
diff --git a/mrbgems/mruby-encoding/src/encoding.c b/mrbgems/mruby-encoding/src/encoding.c
new file mode 100644
index 000000000..cea66d4f0
--- /dev/null
+++ b/mrbgems/mruby-encoding/src/encoding.c
@@ -0,0 +1,133 @@
+#include <mruby.h>
+#include <mruby/string.h>
+#include <mruby/variable.h>
+#include <mruby/internal.h>
+#include <mruby/presym.h>
+
+#define ENC_ASCII_8BIT "ASCII-8BIT"
+#define ENC_BINARY     "BINARY"
+#define ENC_UTF8       "UTF-8"
+
+#define ENC_COMP_P(enc, enc_lit) \
+  casecmp_p(RSTRING_PTR(enc), RSTRING_LEN(enc), enc_lit, sizeof(enc_lit"")-1)
+
+static mrb_bool
+casecmp_p(const char *s1, mrb_int len1, const char *s2, mrb_int len2)
+{
+  if (len1 != len2) return FALSE;
+
+  const char *e1 = s1 + len1;
+  const char *e2 = s2 + len2;
+  while (s1 < e1 && s2 < e2) {
+    if (*s1 != *s2 && TOUPPER(*s1) != TOUPPER(*s2)) return FALSE;
+    s1++;
+    s2++;
+  }
+  return TRUE;
+}
+
+/*
+ * call-seq:
+ *   string.valid_encoding? -> true or false
+ *
+ * Returns true for a string which is encoded correctly.
+ *
+ */
+static mrb_value
+str_valid_enc_p(mrb_state *mrb, mrb_value str)
+{
+#define utf8_islead(c) ((unsigned char)((c)&0xc0) != 0x80)
+
+  struct RString *s = mrb_str_ptr(str);
+  if (RSTR_SINGLE_BYTE_P(s)) return mrb_true_value();
+  if (RSTR_BINARY_P(s)) return mrb_true_value();
+
+  mrb_int byte_len = RSTR_LEN(s);
+  mrb_int utf8_len = 0;
+  const char *p = RSTR_PTR(s);
+  const char *e = p + byte_len;
+  while (p < e) {
+    mrb_int len = mrb_utf8len(p, e);
+
+    if (len == 1 && (*p & 0x80)) return mrb_false_value();
+    p += len;
+    utf8_len++;
+  }
+  if (byte_len == utf8_len) RSTR_SET_SINGLE_BYTE_FLAG(s);
+  return mrb_true_value();
+}
+
+static mrb_value
+str_b(mrb_state *mrb, mrb_value self)
+{
+  mrb_value str = mrb_str_dup(mrb, self);
+  mrb_str_ptr(str)->flags |= MRB_STR_BINARY;
+  return str;
+}
+
+static mrb_value
+get_encoding(mrb_state *mrb, mrb_sym enc)
+{
+  struct RClass *e = mrb_module_get_id(mrb, MRB_SYM(Encoding));
+  return mrb_const_get(mrb, mrb_obj_value(e), enc);
+}
+
+static mrb_value
+str_encoding(mrb_state *mrb, mrb_value self)
+{
+  struct RString *s = mrb_str_ptr(self);
+  if (RSTR_BINARY_P(s)) {
+    return get_encoding(mrb, MRB_SYM(BINARY));
+  }
+  return get_encoding(mrb, MRB_SYM(UTF_8));
+}
+
+static mrb_value
+str_force_encoding(mrb_state *mrb, mrb_value self)
+{
+  mrb_value enc;
+
+  mrb_get_args(mrb, "S", &enc);
+
+  struct RString *s = mrb_str_ptr(self);
+  if (ENC_COMP_P(enc, ENC_ASCII_8BIT) ||
+      ENC_COMP_P(enc, ENC_BINARY)) {
+    s->flags |= MRB_STR_BINARY;
+  }
+  else if (ENC_COMP_P(enc, ENC_UTF8)) {
+    s->flags &= ~MRB_STR_BINARY;
+  }
+  else {
+    mrb_raisef(mrb, E_ARGUMENT_ERROR, "unknown encoding name - %v", enc);
+  }
+  return self;
+}
+
+void
+mrb_mruby_encoding_gem_init(mrb_state* mrb)
+{
+  struct RClass *s = mrb->string_class;
+
+  mrb_define_method_id(mrb, s, MRB_SYM(b),                str_b,               MRB_ARGS_NONE());
+  mrb_define_method_id(mrb, s, MRB_SYM_Q(valid_encoding), str_valid_enc_p,     MRB_ARGS_NONE());
+  mrb_define_method_id(mrb, s, MRB_SYM(encoding),         str_encoding,        MRB_ARGS_NONE());
+  mrb_define_method_id(mrb, s, MRB_SYM(force_encoding),   str_force_encoding,  MRB_ARGS_REQ(1));
+
+  /* Poorman's Encoding
+   *
+   * Encoding - module instead of class
+   * encodings - supports only UTF-8 and ASCII-8BIT (and its alias BINARY)
+   * each Encoding - encoding name string instead of Encoding object
+   *
+   */
+  struct RClass *e = mrb_define_module_id(mrb, MRB_SYM(Encoding));
+  mrb_value b = mrb_str_new_lit_frozen(mrb, ENC_ASCII_8BIT);
+  mrb_define_const_id(mrb, e, MRB_SYM(ASCII_8BIT), b);
+  mrb_define_const_id(mrb, e, MRB_SYM(BINARY), b);
+  mrb_define_const_id(mrb, e, MRB_SYM(UTF_8), mrb_str_new_lit_frozen(mrb, ENC_UTF8));
+}
+
+void
+mrb_mruby_encoding_gem_final(mrb_state* mrb)
+{
+}
diff --git a/mrbgems/mruby-encoding/test/numeric.rb b/mrbgems/mruby-encoding/test/numeric.rb
new file mode 100644
index 000000000..f79235de0
--- /dev/null
+++ b/mrbgems/mruby-encoding/test/numeric.rb
@@ -0,0 +1,27 @@
+assert('Integer#chr') do
+  assert_equal("A", 65.chr)
+  assert_equal("B", 0x42.chr)
+  assert_equal("\xab", 171.chr)
+  assert_raise(RangeError) { -1.chr }
+  assert_raise(RangeError) { 256.chr }
+
+  assert_equal("A", 65.chr("ASCII-8BIT"))
+  assert_equal("B", 0x42.chr("BINARY"))
+  assert_equal("\xab", 171.chr("ascii-8bit"))
+  assert_raise(RangeError) { -1.chr("binary") }
+  assert_raise(RangeError) { 256.chr("Ascii-8bit") }
+  assert_raise(ArgumentError) { 65.chr("ASCII") }
+  assert_raise(ArgumentError) { 65.chr("ASCII-8BIT", 2) }
+  assert_raise(TypeError) { 65.chr(:BINARY) }
+
+  if __ENCODING__ == "ASCII-8BIT"
+    assert_raise(ArgumentError) { 65.chr("UTF-8") }
+  else
+    assert_equal("A", 65.chr("UTF-8"))
+    assert_equal("B", 0x42.chr("UTF-8"))
+    assert_equal("«", 171.chr("utf-8"))
+    assert_equal("あ", 12354.chr("Utf-8"))
+    assert_raise(RangeError) { -1.chr("utf-8") }
+    assert_raise(RangeError) { 0x110000.chr.chr("UTF-8") }
+  end
+end
diff --git a/mrbgems/mruby-encoding/test/string.rb b/mrbgems/mruby-encoding/test/string.rb
new file mode 100644
index 000000000..0c54e6dcf
--- /dev/null
+++ b/mrbgems/mruby-encoding/test/string.rb
@@ -0,0 +1,30 @@
+##
+# String(Ext) Test
+
+UTF8STRING = __ENCODING__ == "UTF-8"
+
+assert('String#valid_encoding?') do
+  assert_true "hello".valid_encoding?
+  if UTF8STRING
+    assert_true "あ".valid_encoding?
+    assert_false "\xfe".valid_encoding?
+    assert_false "あ\xfe".valid_encoding?
+    assert_true "あ\xfe".b.valid_encoding?
+  else
+    assert_true "\xfe".valid_encoding?
+  end
+end
+
+assert('String#encoding') do
+  if UTF8STRING
+    a = "あ"
+    assert_equal Encoding::UTF_8, a.encoding
+    assert_equal Encoding::BINARY, a.b.encoding
+    assert_equal a, a.force_encoding(Encoding::BINARY)
+    assert_equal a, a.force_encoding(Encoding::BINARY)
+    assert_equal Encoding::BINARY, a.encoding
+  else
+    a = "hello"
+    assert_equal Encoding::BINARY, a.encoding
+  end
+end