diff --git a/libraries/grpc-robot/grpc_robot/tools/__init__.py b/libraries/grpc-robot/grpc_robot/tools/__init__.py
new file mode 100644
index 0000000..e69de29
--- /dev/null
+++ b/libraries/grpc-robot/grpc_robot/tools/__init__.py
diff --git a/libraries/grpc-robot/grpc_robot/tools/dmi_tools.py b/libraries/grpc-robot/grpc_robot/tools/dmi_tools.py
new file mode 100644
index 0000000..c47191c
--- /dev/null
+++ b/libraries/grpc-robot/grpc_robot/tools/dmi_tools.py
@@ -0,0 +1,75 @@
+# Copyright 2020 ADTRAN, Inc.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+from grpc_robot.grpc_robot import _package_version_get
+
+from dmi import hw_metrics_mgmt_service_pb2, hw_events_mgmt_service_pb2
+from ..tools.protobuf_to_dict import protobuf_to_dict
+
+
+class DmiTools(object):
+    """
+    Tools for the device-management-interface, e.g decoding / conversions.
+    """
+
+    try:
+        ROBOT_LIBRARY_VERSION = _package_version_get('grpc_robot')
+    except NameError:
+        ROBOT_LIBRARY_VERSION = 'unknown'
+
+    @staticmethod
+    def hw_metrics_mgmt_decode_metric(bytestring, return_enum_integer='false', return_defaults='false'):
+        """
+        Converts bytes to a Metric as defined in _message Metric_ from hw_metrics_mgmt_service.proto
+
+        *Parameters*:
+        - bytestring: <bytes>; Byte string, e.g. as it comes from Kafka messages.
+        - return_enum_integer: <bool> or <string>; Whether or not to return the enum values as integer values rather than their labels. Default: ${FALSE} or false.
+        - return_defaults: <bool> or <string>; Whether or not to return the default values. Default: _${FALSE}_ or _false_.
+
+        *Return*: A dictionary with same structure as the _metric_ key from the return dictionary of keyword _Hw Metrics Mgmt Service Get Metric_.
+
+        *Example*:
+        | Import Library | grpc_robot.DmiTools | WITH NAME | dmi_tools |
+        | ${kafka_records} | kafka.Records Get |
+        | FOR | ${kafka_record} | IN | @{kafka_records} |
+        |  | ${metric} | dmi_tools.Hw Metrics Mgmt Decode Metric | ${kafka_record}[message] |
+        |  | Log | ${metric} |
+        | END |
+        """
+        return_enum_integer = bool(str(return_enum_integer).lower() == 'true')
+        metric = hw_metrics_mgmt_service_pb2.Metric.FromString(bytestring)
+        return protobuf_to_dict(metric, use_enum_labels=not return_enum_integer, including_default_value_fields=return_defaults)
+
+    @staticmethod
+    def hw_events_mgmt_decode_event(bytestring, return_enum_integer='false', return_defaults='false'):
+        """
+        Converts bytes to a Event as defined in _message Event_ from hw_events_mgmt_service.proto
+
+        *Parameters*:
+        - bytestring: <bytes>; Byte string, e.g. as it comes from Kafka messages.
+        - return_enum_integer: <bool> or <string>; Whether or not to return the enum values as integer values rather than their labels. Default: ${FALSE} or false.
+        - return_defaults: <bool> or <string>; Whether or not to return the default values. Default: _${FALSE}_ or _false_.
+
+        *Return*: A dictionary with same structure as the _event_ key from the return dictionary of keyword _Hw Event Mgmt Service List Events_.
+
+        *Example*:
+        | Import Library | grpc_robot.DmiTools | WITH NAME | dmi_tools |
+        | ${kafka_records} | kafka.Records Get |
+        | FOR | ${kafka_record} | IN | @{kafka_records} |
+        |  | ${metric} | dmi_tools.Hw Events Mgmt Decode Event | ${kafka_record}[message] |
+        |  | Log | ${metric} |
+        | END |
+        """
+        return_enum_integer = bool(str(return_enum_integer).lower() == 'true')
+        event = hw_events_mgmt_service_pb2.Event.FromString(bytestring)
+        return protobuf_to_dict(event, use_enum_labels=not return_enum_integer, including_default_value_fields=return_defaults)
diff --git a/libraries/grpc-robot/grpc_robot/tools/protobuf_parse.py b/libraries/grpc-robot/grpc_robot/tools/protobuf_parse.py
new file mode 100644
index 0000000..c850d8c
--- /dev/null
+++ b/libraries/grpc-robot/grpc_robot/tools/protobuf_parse.py
@@ -0,0 +1,445 @@
+# kann keine Kommentare
+
+# -*- coding: utf-8 -*-
+
+# Parser for protocol buffer .proto files
+import enum as stdlib_enum
+from string import ascii_letters, digits, hexdigits, octdigits
+
+import attr
+
+from parsy import char_from, from_enum, generate, regex, seq, string
+
+# This file follows the spec at
+# https://developers.google.com/protocol-buffers/docs/reference/proto3-spec
+# very closely.
+
+# However, because we are parsing into useful objects, we do transformations
+# along the way e.g. turning into integers, strings etc. and custom objects.
+# Some of the lowest level items have been implemented using 'regex' and converting
+# the descriptions to regular expressions. Higher level constructs have been
+# implemented using other parsy primitives and combinators.
+
+# Notes:
+
+# 1. Whitespace is very badly defined in the 'spec', so we guess what is meant.
+# 2. The spec doesn't allow for comments, and neither does this parser.
+#    Other places mention that C++ style comments are allowed. To support that,
+#    this parser would need to be changed into split lexing/parsing stages
+#    (otherwise you hit issues with comments start markers within string literals).
+# 3. Other notes inline.
+
+
+# Our utilities
+optional_string = lambda s: string(s).times(0, 1).concat()
+convert_decimal = int
+convert_octal = lambda s: int(s, 8)
+convert_hex = lambda s: int(s, 16)
+exclude_none = lambda l: [i for i in l if i is not None]
+
+
+def lexeme(p):
+    """
+    From a parser (or string), make a parser that consumes
+    whitespace on either side.
+    """
+    if isinstance(p, str):
+        p = string(p)
+    return regex(r'\s*') >> p << regex(r'\s*')
+
+
+def is_present(p):
+    """
+    Given a parser or string, make a parser that returns
+    True if the parser matches, False otherwise
+    """
+    return lexeme(p).optional().map(lambda v: False if v is None else True)
+
+
+# Our data structures
+@attr.s
+class Import:
+    identifier = attr.ib()
+    option = attr.ib()
+
+
+@attr.s
+class Package:
+    identifer = attr.ib()
+
+
+@attr.s
+class Option:
+    name = attr.ib()
+    value = attr.ib()
+
+
+@attr.s
+class Field:
+    repeated = attr.ib()
+    type = attr.ib()
+    name = attr.ib()
+    number = attr.ib()
+    options = attr.ib()
+
+
+@attr.s
+class OneOfField:
+    type = attr.ib()
+    name = attr.ib()
+    number = attr.ib()
+    options = attr.ib()
+
+
+@attr.s
+class OneOf:
+    name = attr.ib()
+    fields = attr.ib()
+
+
+@attr.s
+class Map:
+    key_type = attr.ib()
+    type = attr.ib()
+    name = attr.ib()
+    number = attr.ib()
+    options = attr.ib()
+
+
+@attr.s
+class Reserved:
+    items = attr.ib()
+
+
+@attr.s
+class Range:
+    from_ = attr.ib()
+    to = attr.ib()
+
+
+@attr.s
+class EnumField:
+    name = attr.ib()
+    value = attr.ib()
+    options = attr.ib()
+
+
+@attr.s
+class Enum:
+    name = attr.ib()
+    body = attr.ib()
+
+
+@attr.s
+class Message:
+    name = attr.ib()
+    body = attr.ib()
+
+
+@attr.s
+class Service:
+    name = attr.ib()
+    body = attr.ib()
+
+
+@attr.s
+class Rpc:
+    name = attr.ib()
+    request_stream = attr.ib()
+    request_message_type = attr.ib()
+    response_stream = attr.ib()
+    response_message_type = attr.ib()
+    options = attr.ib()
+
+
+@attr.s
+class Proto:
+    syntax = attr.ib()
+    statements = attr.ib()
+
+
+# Enums:
+class ImportOption(stdlib_enum.Enum):
+    WEAK = "weak"
+    PUBLIC = "public"
+
+
+class Type(stdlib_enum.Enum):
+    DOUBLE = "double"
+    FLOAT = "float"
+    INT32 = "int32"
+    INT64 = "int64"
+    UINT32 = "uint32"
+    UINT64 = "uint64"
+    SINT32 = "sint32"
+    SINT64 = "sint64"
+    FIXED32 = "fixed32"
+    FIXED64 = "fixed64"
+    SFIXED32 = "sfixed32"
+    SFIXED64 = "sfixed64"
+    BOOL = "bool"
+    STRING = "string"
+    BYTES = "bytes"
+
+
+class KeyType(stdlib_enum.Enum):
+    INT32 = "int32"
+    INT64 = "int64"
+    UINT32 = "uint32"
+    UINT64 = "uint64"
+    SINT32 = "sint32"
+    SINT64 = "sint64"
+    FIXED32 = "fixed32"
+    FIXED64 = "fixed64"
+    SFIXED32 = "sfixed32"
+    SFIXED64 = "sfixed64"
+    BOOL = "bool"
+    STRING = "string"
+
+
+# Some extra constants to avoid typing
+SEMI, EQ, LPAREN, RPAREN, LBRACE, RBRACE, LBRAC, RBRAC = [lexeme(c) for c in ";=(){}[]"]
+
+
+# -- Beginning of following spec --
+# Letters and digits
+letter = char_from(ascii_letters)
+decimalDigit = char_from(digits)
+octalDigit = char_from(octdigits)
+hexDigit = char_from(hexdigits)
+
+# Identifiers
+
+# Compared to spec, we add some '_' prefixed items which are not wrapped in `lexeme`,
+# on the assumption that spaces in the middle of identifiers are not accepted.
+_ident = (letter + (letter | decimalDigit | string("_")).many().concat()).desc('ident')
+ident = lexeme(_ident)
+fullIdent = lexeme(ident + (string(".") + ident).many().concat()).desc('fullIdent')
+_messageName = _ident
+messageName = lexeme(ident).desc('messageName')
+_enumName = ident
+enumName = lexeme(_enumName).desc('enumName')
+fieldName = ident.desc('fieldName')
+oneofName = ident.desc('oneofName')
+mapName = ident.desc('mapName')
+serviceName = ident.desc('serviceName')
+rpcName = ident.desc('rpcName')
+messageType = optional_string(".") + (_ident + string(".")).many().concat() + _messageName
+enumType = optional_string(".") + (_ident + string(".")).many().concat() + _enumName
+
+# Integer literals
+decimalLit = regex("[1-9][0-9]*").desc('decimalLit').map(convert_decimal)
+octalLit = regex("0[0-7]*").desc('octalLit').map(convert_octal)
+hexLit = regex("0[x|X][0-9a-fA-F]+").desc('octalLit').map(convert_hex)
+intLit     = decimalLit | octalLit | hexLit
+
+
+# Floating-point literals
+decimals = r'[0-9]+'
+exponent = r'[e|E][+|-]?' + decimals
+floatLit = regex(r'({decimals}\.({decimals})?({exponent})?)|{decimals}{exponent}|\.{decimals}({exponent})?'
+                 .format(decimals=decimals, exponent=exponent)).desc('floatLit').map(float)
+
+
+# Boolean
+boolLit = (string("true").result(True) | string("false").result(False)).desc('boolLit')
+
+
+# String literals
+hexEscape = regex(r"\\[x|X]") >> regex("[0-9a-fA-F]{2}").map(convert_hex).map(chr)
+octEscape = regex(r"\\") >> regex('[0-7]{2}').map(convert_octal).map(chr)
+charEscape = regex(r"\\") >> (
+    string("a").result("\a")
+    | string("b").result("\b")
+    | string("f").result("\f")
+    | string("n").result("\n")
+    | string("r").result("\r")
+    | string("t").result("\t")
+    | string("v").result("\v")
+    | string("\\").result("\\")
+    | string("'").result("'")
+    | string('"').result('"')
+)
+escapes = hexEscape | octEscape | charEscape
+# Correction to spec regarding " and ' inside quoted strings
+strLit = (string("'") >> (escapes | regex(r"[^\0\n\'\\]")).many().concat() << string("'")
+          | string('"') >> (escapes | regex(r"[^\0\n\"\\]")).many().concat() << string('"')).desc('strLit')
+quote = string("'") | string('"')
+
+# EmptyStatement
+emptyStatement = string(";").result(None)
+
+# Signed numbers:
+# (Extra compared to spec, to cope with need to produce signed numeric values)
+signedNumberChange = lambda s, num: (-1) if s == "-" else (+1)
+sign = regex("[-+]?")
+signedIntLit = seq(sign, intLit).combine(signedNumberChange)
+signedFloatLit = seq(sign, floatLit).combine(signedNumberChange)
+
+
+# Constant
+# put fullIdent at end to disabmiguate from boolLit
+constant = signedIntLit | signedFloatLit | strLit | boolLit | fullIdent
+
+# Syntax
+syntax = lexeme("syntax") >> EQ >> quote >> string("proto3") << quote + SEMI
+
+# Import Statement
+import_option = from_enum(ImportOption)
+
+import_ = seq(lexeme("import") >> import_option.optional().tag('option'),
+              lexeme(strLit).tag('identifier') << SEMI).combine_dict(Import)
+
+# Package
+package = seq(lexeme("package") >> fullIdent << SEMI).map(Package)
+
+# Option
+optionName = (ident | (LPAREN >> fullIdent << RPAREN)) + (string(".") + ident).many().concat()
+option = seq(lexeme("option") >> optionName.tag('name'),
+             EQ >> constant.tag('value') << SEMI,
+             ).combine_dict(Option)
+
+# Normal field
+type_ = lexeme(from_enum(Type) | messageType | enumType)
+fieldNumber = lexeme(intLit)
+
+fieldOption = seq(optionName.tag('name'),
+                  EQ >> constant.tag('value')).combine_dict(Option)
+fieldOptions = fieldOption.sep_by(lexeme(","), min=1)
+fieldOptionList = (lexeme("[") >> fieldOptions << lexeme("]")).optional().map(
+    lambda o: [] if o is None else o)
+
+field = seq(is_present("repeated").tag('repeated'),
+            type_.tag('type'),
+            fieldName.tag('name') << EQ,
+            fieldNumber.tag('number'),
+            fieldOptionList.tag('options') << SEMI,
+            ).combine_dict(Field)
+
+# Oneof and oneof field
+oneofField = seq(type_.tag('type'),
+                 fieldName.tag('name') << EQ,
+                 fieldNumber.tag('number'),
+                 fieldOptionList.tag('options') << SEMI,
+                 ).combine_dict(OneOfField)
+oneof = seq(lexeme("oneof") >> oneofName.tag('name'),
+            LBRACE
+            >> (oneofField | emptyStatement).many().map(exclude_none).tag('fields')
+            << RBRACE
+            ).combine_dict(OneOf)
+
+# Map field
+keyType = lexeme(from_enum(KeyType))
+mapField = seq(lexeme("map") >> lexeme("<") >> keyType.tag('key_type'),
+               lexeme(",") >> type_.tag('type'),
+               lexeme(">") >> mapName.tag('name'),
+               EQ >> fieldNumber.tag('number'),
+               fieldOptionList.tag('options') << SEMI
+               ).combine_dict(Map)
+
+# Reserved
+range_ = seq(lexeme(intLit).tag('from_'),
+             (lexeme("to") >> (intLit | lexeme("max"))).optional().tag('to')
+             ).combine_dict(Range)
+ranges = range_.sep_by(lexeme(","), min=1)
+# The spec for 'reserved' indicates 'fieldName' here, which is never a quoted string.
+# But the example has a quoted string. We have changed it to 'strLit'
+fieldNames = strLit.sep_by(lexeme(","), min=1)
+reserved = seq(lexeme("reserved") >> (ranges | fieldNames) << SEMI
+               ).combine(Reserved)
+
+# Enum definition
+enumValueOption = seq(optionName.tag('name') << EQ,
+                      constant.tag('value')
+                      ).combine_dict(Option)
+enumField = seq(ident.tag('name') << EQ,
+                lexeme(intLit).tag('value'),
+                (lexeme("[") >> enumValueOption.sep_by(lexeme(","), min=1) << lexeme("]")).optional()
+                .map(lambda o: [] if o is None else o).tag('options')
+                << SEMI
+                ).combine_dict(EnumField)
+enumBody = (LBRACE
+            >> (option | enumField | emptyStatement).many().map(exclude_none)
+            << RBRACE)
+enum = seq(lexeme("enum") >> enumName.tag('name'),
+           enumBody.tag('body')
+           ).combine_dict(Enum)
+
+
+# Message definition
+@generate
+def message():
+    yield lexeme("message")
+    name = yield messageName
+    body = yield messageBody
+    return Message(name=name, body=body)
+
+
+messageBody = (LBRACE
+               >> (field | enum | message | option | oneof | mapField
+                   | reserved | emptyStatement).many()
+               << RBRACE)
+
+
+# Service definition
+rpc = seq(lexeme("rpc") >> rpcName.tag('name'),
+          LPAREN
+          >> (is_present("stream").tag("request_stream")),
+          messageType.tag("request_message_type") << RPAREN,
+          lexeme("returns") >> LPAREN
+          >> (is_present("stream").tag("response_stream")),
+          messageType.tag("response_message_type")
+          << RPAREN,
+          ((LBRACE
+           >> (option | emptyStatement).many()
+           << RBRACE)
+           | SEMI.result([])
+           ).optional().map(exclude_none).tag('options')
+          ).combine_dict(Rpc)
+
+service = seq(lexeme("service") >> serviceName.tag('name'),
+              LBRACE
+              >> (option | rpc | emptyStatement).many().map(exclude_none).tag('body')
+              << RBRACE
+              ).combine_dict(Service)
+
+
+# Proto file
+topLevelDef = message | enum | service
+proto = seq(syntax.tag('syntax'),
+            (import_ | package | option | topLevelDef | emptyStatement
+             ).many().map(exclude_none).tag('statements')
+            ).combine_dict(Proto)
+
+
+EXAMPLE = """syntax = "proto3";
+import public "other.proto";
+option java_package = "com.example.foo";
+option java_package = "com.example.foo";
+package dmi;
+
+enum EnumAllowingAlias {
+  option allow_alias = true;
+  UNKNOWN = 0;
+  STARTED = 1;
+  RUNNING = 2 [(custom_option) = "hello world"];
+}
+message outer {
+  option (my_option).a = true;
+  message inner {
+    int64 ival = 1;
+  }
+  repeated inner inner_message = 2;
+  EnumAllowingAlias enum_field =3;
+  map<int32, string> my_map = 4;
+  oneof operation {
+    MetricsConfig changes = 2;
+    bool reset_to_default = 3;
+  }
+}
+"""
+# Smoke test - should find 4 top level statements in the example:
+# assert len(proto.parse(EXAMPLE).statements) == 4
+# print(proto.parse(EXAMPLE).statements)
+# for st in proto.parse(EXAMPLE).statements:
+#     print(type(st))
diff --git a/libraries/grpc-robot/grpc_robot/tools/protobuf_to_dict.py b/libraries/grpc-robot/grpc_robot/tools/protobuf_to_dict.py
new file mode 100644
index 0000000..bf4538a
--- /dev/null
+++ b/libraries/grpc-robot/grpc_robot/tools/protobuf_to_dict.py
@@ -0,0 +1,227 @@
+# -*- coding:utf-8 -*-
+
+# copied from https://github.com/kaporzhu/protobuf-to-dict
+# all credits to this script go to Kapor Zhu (kapor.zhu@gmail.com)
+#
+# need a fix for bug: "Use enum_label when setting the default value if use_enum_labels is true" (line 95)
+
+import base64
+
+import six
+
+from google.protobuf.message import Message
+from google.protobuf.descriptor import FieldDescriptor
+
+
+__all__ = ["protobuf_to_dict", "TYPE_CALLABLE_MAP", "dict_to_protobuf",
+           "REVERSE_TYPE_CALLABLE_MAP"]
+
+
+EXTENSION_CONTAINER = '___X'
+
+
+TYPE_CALLABLE_MAP = {
+    FieldDescriptor.TYPE_DOUBLE: float,
+    FieldDescriptor.TYPE_FLOAT: float,
+    FieldDescriptor.TYPE_INT32: int,
+    FieldDescriptor.TYPE_INT64: int if six.PY3 else six.integer_types[1],
+    FieldDescriptor.TYPE_UINT32: int,
+    FieldDescriptor.TYPE_UINT64: int if six.PY3 else six.integer_types[1],
+    FieldDescriptor.TYPE_SINT32: int,
+    FieldDescriptor.TYPE_SINT64: int if six.PY3 else six.integer_types[1],
+    FieldDescriptor.TYPE_FIXED32: int,
+    FieldDescriptor.TYPE_FIXED64: int if six.PY3 else six.integer_types[1],
+    FieldDescriptor.TYPE_SFIXED32: int,
+    FieldDescriptor.TYPE_SFIXED64: int if six.PY3 else six.integer_types[1],
+    FieldDescriptor.TYPE_BOOL: bool,
+    FieldDescriptor.TYPE_STRING: six.text_type,
+    FieldDescriptor.TYPE_BYTES: six.binary_type,
+    FieldDescriptor.TYPE_ENUM: int,
+}
+
+
+def repeated(type_callable):
+    return lambda value_list: [type_callable(value) for value in value_list]
+
+
+def enum_label_name(field, value):
+    return field.enum_type.values_by_number[int(value)].name
+
+
+def _is_map_entry(field):
+    return (field.type == FieldDescriptor.TYPE_MESSAGE and
+            field.message_type.has_options and
+            field.message_type.GetOptions().map_entry)
+
+
+def protobuf_to_dict(pb, type_callable_map=TYPE_CALLABLE_MAP, use_enum_labels=False,
+                     including_default_value_fields=False):
+    result_dict = {}
+    extensions = {}
+    for field, value in pb.ListFields():
+        if field.message_type and field.message_type.has_options and field.message_type.GetOptions().map_entry:
+            result_dict[field.name] = dict()
+            value_field = field.message_type.fields_by_name['value']
+            type_callable = _get_field_value_adaptor(
+                pb, value_field, type_callable_map,
+                use_enum_labels, including_default_value_fields)
+            for k, v in value.items():
+                result_dict[field.name][k] = type_callable(v)
+            continue
+        type_callable = _get_field_value_adaptor(pb, field, type_callable_map,
+                                                 use_enum_labels, including_default_value_fields)
+        if field.label == FieldDescriptor.LABEL_REPEATED:
+            type_callable = repeated(type_callable)
+
+        if field.is_extension:
+            extensions[str(field.number)] = type_callable(value)
+            continue
+
+        result_dict[field.name] = type_callable(value)
+
+    # Serialize default value if including_default_value_fields is True.
+    if including_default_value_fields:
+        for field in pb.DESCRIPTOR.fields:
+            # Singular message fields and oneof fields will not be affected.
+            if ((
+                    field.label != FieldDescriptor.LABEL_REPEATED and
+                    field.cpp_type == FieldDescriptor.CPPTYPE_MESSAGE) or
+                    field.containing_oneof):
+                continue
+            if field.name in result_dict:
+                # Skip the field which has been serailized already.
+                continue
+            if _is_map_entry(field):
+                result_dict[field.name] = {}
+            else:
+                if use_enum_labels and field.type == FieldDescriptor.TYPE_ENUM:
+                    result_dict[field.name] = enum_label_name(field, field.default_value)
+                else:
+                    result_dict[field.name] = field.default_value
+
+    if extensions:
+        result_dict[EXTENSION_CONTAINER] = extensions
+    return result_dict
+
+
+def _get_field_value_adaptor(pb, field, type_callable_map=TYPE_CALLABLE_MAP, use_enum_labels=False,
+                             including_default_value_fields=False):
+    if field.type == FieldDescriptor.TYPE_MESSAGE:
+        # recursively encode protobuf sub-message
+        return lambda pb: protobuf_to_dict(
+            pb, type_callable_map=type_callable_map,
+            use_enum_labels=use_enum_labels,
+            including_default_value_fields=including_default_value_fields,
+        )
+
+    if use_enum_labels and field.type == FieldDescriptor.TYPE_ENUM:
+        return lambda value: enum_label_name(field, value)
+
+    if field.type in type_callable_map:
+        return type_callable_map[field.type]
+
+    raise TypeError("Field %s.%s has unrecognised type id %d" % (
+        pb.__class__.__name__, field.name, field.type))
+
+
+REVERSE_TYPE_CALLABLE_MAP = {
+}
+
+
+def dict_to_protobuf(pb_klass_or_instance, values, type_callable_map=REVERSE_TYPE_CALLABLE_MAP, strict=True, ignore_none=False):
+    """Populates a protobuf model from a dictionary.
+
+    :param pb_klass_or_instance: a protobuf message class, or an protobuf instance
+    :type pb_klass_or_instance: a type or instance of a subclass of google.protobuf.message.Message
+    :param dict values: a dictionary of values. Repeated and nested values are
+       fully supported.
+    :param dict type_callable_map: a mapping of protobuf types to callables for setting
+       values on the target instance.
+    :param bool strict: complain if keys in the map are not fields on the message.
+    :param bool strict: ignore None-values of fields, treat them as empty field
+    """
+    if isinstance(pb_klass_or_instance, Message):
+        instance = pb_klass_or_instance
+    else:
+        instance = pb_klass_or_instance()
+    return _dict_to_protobuf(instance, values, type_callable_map, strict, ignore_none)
+
+
+def _get_field_mapping(pb, dict_value, strict):
+    field_mapping = []
+    for key, value in dict_value.items():
+        if key == EXTENSION_CONTAINER:
+            continue
+        if key not in pb.DESCRIPTOR.fields_by_name:
+            if strict:
+                raise KeyError("%s does not have a field called %s" % (pb, key))
+            continue
+        field_mapping.append((pb.DESCRIPTOR.fields_by_name[key], value, getattr(pb, key, None)))
+
+    for ext_num, ext_val in dict_value.get(EXTENSION_CONTAINER, {}).items():
+        try:
+            ext_num = int(ext_num)
+        except ValueError:
+            raise ValueError("Extension keys must be integers.")
+        if ext_num not in pb._extensions_by_number:
+            if strict:
+                raise KeyError("%s does not have a extension with number %s. Perhaps you forgot to import it?" % (pb, key))
+            continue
+        ext_field = pb._extensions_by_number[ext_num]
+        pb_val = None
+        pb_val = pb.Extensions[ext_field]
+        field_mapping.append((ext_field, ext_val, pb_val))
+
+    return field_mapping
+
+
+def _dict_to_protobuf(pb, value, type_callable_map, strict, ignore_none):
+    fields = _get_field_mapping(pb, value, strict)
+
+    for field, input_value, pb_value in fields:
+        if ignore_none and input_value is None:
+            continue
+        if field.label == FieldDescriptor.LABEL_REPEATED:
+            if field.message_type and field.message_type.has_options and field.message_type.GetOptions().map_entry:
+                value_field = field.message_type.fields_by_name['value']
+                for key, value in input_value.items():
+                    if value_field.cpp_type == FieldDescriptor.CPPTYPE_MESSAGE:
+                        _dict_to_protobuf(getattr(pb, field.name)[key], value, type_callable_map, strict, ignore_none)
+                    else:
+                        getattr(pb, field.name)[key] = value
+                continue
+            for item in input_value:
+                if field.type == FieldDescriptor.TYPE_MESSAGE:
+                    m = pb_value.add()
+                    _dict_to_protobuf(m, item, type_callable_map, strict, ignore_none)
+                elif field.type == FieldDescriptor.TYPE_ENUM and isinstance(item, six.string_types):
+                    pb_value.append(_string_to_enum(field, item))
+                else:
+                    pb_value.append(item)
+            continue
+        if field.type == FieldDescriptor.TYPE_MESSAGE:
+            _dict_to_protobuf(pb_value, input_value, type_callable_map, strict, ignore_none)
+            continue
+
+        if field.type in type_callable_map:
+            input_value = type_callable_map[field.type](input_value)
+
+        if field.is_extension:
+            pb.Extensions[field] = input_value
+            continue
+
+        if field.type == FieldDescriptor.TYPE_ENUM and isinstance(input_value, six.string_types):
+            input_value = _string_to_enum(field, input_value)
+
+        setattr(pb, field.name, input_value)
+
+    return pb
+
+
+def _string_to_enum(field, input_value):
+    enum_dict = field.enum_type.values_by_name
+    try:
+        input_value = enum_dict[input_value].number
+    except KeyError:
+        raise KeyError("`%s` is not a valid value for field `%s`" % (input_value, field.name))
+    return input_value
diff --git a/libraries/grpc-robot/grpc_robot/tools/protop.py b/libraries/grpc-robot/grpc_robot/tools/protop.py
new file mode 100644
index 0000000..0e0f00c
--- /dev/null
+++ b/libraries/grpc-robot/grpc_robot/tools/protop.py
@@ -0,0 +1,191 @@
+# Copyright 2020 ADTRAN, Inc.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+import os
+import re
+import glob
+import json
+import argparse
+
+from . import protobuf_parse as parser
+
+__version__ = '1.0'
+
+USAGE = """ProtoBuf -- Parser of Protocol Buffer to create the input JSON file for the library
+
+Usage:  grpc_robot.protop [options] target_version
+
+ProtoBuf parser can be used to parse ProtoBuf files (*.proto) into a json formatted input file 
+for the grpc_robot library to be used for keyword documentation.
+
+"""
+
+EPILOG = """
+Example
+=======
+# Executing `grpc_robot.protop` module using Python.
+$ grpc_robot.protop -i /home/user/Workspace/grpc/proto/dmi 0.9.1
+"""
+
+
+class ProtoBufParser(object):
+
+    def __init__(self, target, target_version, input_dir, output_dir=None):
+
+        super().__init__()
+
+        self.target = target
+        self.target_version = target_version.replace('.', '_')
+        self.input_dir = input_dir
+        self.output_dir = output_dir
+
+    @staticmethod
+    def read_enum(enum, protobuf_dict, module):
+        enum_dict = {'name': enum.name, 'type': 'enum', 'module': module, 'values': {ef.value: ef.name for ef in enum.body}}
+        protobuf_dict['data_types'].append(enum_dict)
+
+    def read_message(self, message, protobuf_dict, module):
+        message_dict = {'name': message.name, 'type': 'message', 'module': module, 'fields': []}
+
+        for f in message.body:
+            if isinstance(f, parser.Enum):
+                self.read_enum(f, protobuf_dict, module)
+                continue
+
+            elif isinstance(f, parser.Message):
+                self.read_message(f, protobuf_dict, module)
+                continue
+
+            field_dict = {'name': f.name, 'is_choice': isinstance(f, parser.OneOf)}
+
+            if isinstance(f, parser.Field):
+                field_dict['repeated'] = f.repeated
+
+                try:
+                    field_dict['type'] = f.type._value_
+                    field_dict['lookup'] = False
+                except AttributeError:
+                    field_dict['type'] = f.type
+                    field_dict['lookup'] = True
+
+            elif isinstance(f, parser.OneOf):
+                field_dict['cases'] = []
+                for c in f.fields:
+                    case_dict = {'name': c.name}
+                    try:
+                        case_dict['type'] = c.type._value_
+                        case_dict['lookup'] = False
+                    except AttributeError:
+                        case_dict['type'] = c.type
+                        case_dict['lookup'] = True
+                    field_dict['cases'].append(case_dict)
+
+            message_dict['fields'].append(field_dict)
+
+        protobuf_dict['data_types'].append(message_dict)
+
+    def parse_files(self):
+
+        protobuf_dict = {
+            'modules': [],
+            'data_types': [],
+            'services': []
+        }
+
+        for file_name in glob.glob(os.path.join(self.input_dir, '*.proto')):
+
+            module = os.path.splitext(os.path.basename(file_name))[0]
+            module_dict = {'name': module, 'imports': []}
+
+            # the protobuf parser can not handle comments "// ...", so remove them first from the file
+            file_content = re.sub(r'\/\/.*', '', open(file_name).read())
+            parsed = parser.proto.parse(file_content)
+
+            for p in parsed.statements:
+
+                if isinstance(p, parser.Import):
+                    module_dict['imports'].append(os.path.splitext(os.path.basename(p.identifier))[0])
+
+                elif isinstance(p, parser.Enum):
+                    self.read_enum(p, protobuf_dict, module)
+
+                elif isinstance(p, parser.Message):
+                    self.read_message(p, protobuf_dict, module)
+
+                elif isinstance(p, parser.Service):
+                    service_dict = {'name': p.name, 'module': module, 'rpcs': []}
+
+                    for field in p.body:
+
+                        if isinstance(field, parser.Enum):
+                            self.read_enum(field, protobuf_dict, module)
+
+                        elif isinstance(field, parser.Message):
+                            self.read_message(field, protobuf_dict, module)
+
+                        elif isinstance(field, parser.Rpc):
+                            rpc_dict = {'name': field.name, 'request': {}, 'response': {}}
+
+                            for attr in ['request', 'response']:
+                                try:
+                                    rpc_dict[attr]['is_stream'] = field.__getattribute__('%s_stream' % attr)
+
+                                    try:
+                                        rpc_dict[attr]['type'] = field.__getattribute__('%s_message_type' % attr)._value_
+                                        rpc_dict[attr]['lookup'] = False
+                                    except AttributeError:
+                                        rpc_dict[attr]['type'] = field.__getattribute__('%s_message_type' % attr)
+                                        rpc_dict[attr]['lookup'] = not rpc_dict[attr]['type'].lower().startswith('google.protobuf.')
+
+                                except AttributeError:
+                                    rpc_dict[attr] = None
+
+                            service_dict['rpcs'].append(rpc_dict)
+
+                    protobuf_dict['services'].append(service_dict)
+
+            protobuf_dict['modules'].append(module_dict)
+
+        if self.output_dir is not None:
+            json_file_name = os.path.join(self.output_dir, self.target, '%s_%s' % (self.target, self.target_version), '%s.json' % self.target)
+            json.dump(protobuf_dict, open(json_file_name, 'w'))
+
+        return protobuf_dict
+
+
+base_dir = os.path.dirname(os.path.realpath(__file__))
+output_base_dir = os.path.join(os.path.split(base_dir)[:-1][0], 'services')
+
+
+def main():
+    # create commandline parser
+    arg_parse = argparse.ArgumentParser(description=USAGE, epilog=EPILOG, formatter_class=argparse.RawTextHelpFormatter)
+
+    # add parser options
+    arg_parse.add_argument('target', choices=['dmi'], default='dmi',
+                           help="Target type of which the ProtocolBuffer files shall be converted to the JSON file.")
+    arg_parse.add_argument('target_version', help="Version number of the ProtocolBuffer files.")
+
+    arg_parse.add_argument('-i', '--inputdir', default=os.getcwd(), help="Path to the location of the ProtocolBuffer files.")
+    arg_parse.add_argument('-o', '--outputdir', default=os.getcwd(), help="Path to the location JSON file to be stored.")
+
+    arg_parse.add_argument('-v', '--version', action='version', version=__version__)
+    arg_parse.set_defaults(feature=False)
+
+    # parse commandline
+    args = arg_parse.parse_args()
+
+    ProtoBufParser(args.target, args.target_version, args.inputdir or os.getcwd(), args.outputdir or output_base_dir).parse_files()
+
+
+if __name__ == '__main__':
+    main()
diff --git a/libraries/grpc-robot/grpc_robot/tools/robot_tools.py b/libraries/grpc-robot/grpc_robot/tools/robot_tools.py
new file mode 100644
index 0000000..798bf17
--- /dev/null
+++ b/libraries/grpc-robot/grpc_robot/tools/robot_tools.py
@@ -0,0 +1,115 @@
+# Copyright 2020 ADTRAN, Inc.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+from grpc_robot.grpc_robot import _package_version_get
+
+
+class Collections(object):
+    """
+    Tools for collections (list, dict) related functionality.
+    """
+
+    try:
+        ROBOT_LIBRARY_VERSION = _package_version_get('grpc_robot')
+    except NameError:
+        ROBOT_LIBRARY_VERSION = 'unknown'
+
+    @staticmethod
+    def dict_get_key_by_value(input_dict, search_value):
+        """
+        Gets the first key from _input_dict_ which has the value of _search_value_.
+
+        If _search_value_ is not found in _input_dict_, an empty string is returned.
+
+        *Parameters*:
+        - _input_dict_: <dictionary> to be browsed.
+        - _search_value_: <string>, value to be searched for.
+
+        *Return*: key of dictionary if search value is in input_dict else empty string
+        """
+        return_key = ''
+        for key, val in input_dict.items():
+            if val == search_value:
+                return_key = key
+                break
+
+        return return_key
+
+    @staticmethod
+    def dict_get_value(values_dict, key, strict=False):
+        """
+        Returns the value for given _key_ in _values_dict_.
+
+        If _strict_ is set to False (default) it will return given _key_ if its is not in the dictionary.
+        If set to True, an AssertionError is raised.
+
+        *Parameters*:
+        - _key_: <string>, key to be searched in dictionary.
+        - _values_dict_: <dictionary> in which the key is searched.
+        - _strict_: Optional: <boolean> switch to indicate if an exception shall be raised if key is not in values_dict.
+                Default: False
+
+        *Return*:
+        - if key is in values_dict: Value from _values_dict_ for _key_.
+        - else: _key_.
+        - raises AssertionError in case _key_ is not in _values_dict_ and _strict_ is True.
+        """
+        try:
+            return_value = values_dict[key]
+        except KeyError:
+            if strict:
+                raise AssertionError('Error: Value not found for key: %s' % key)
+            else:
+                return_value = key
+
+        return return_value
+
+    @staticmethod
+    def list_get_dict_by_value(input_list, key_name, value, match='first'):
+        """
+        Retrieves a dictionary from a list of dictionaries where _key_name_ has the _value, if _match_ is
+        "first". Else it returns all matching dictionaries.
+
+        *Parameters*:
+        - _input_list_: <list> ; List of dictionaries.
+        - _key_name_: <dictionary> or <list> ; Name of the key to be searched for.
+        - _value_: <string> or <number> ; Any value of key _key_name_ to be searched for.
+
+        *Example*:
+        | ${dict1}    | Create Dictionary      | key_key=master1 | key1=value11 | key2=value12 |          |
+        | ${dict2}    | Create Dictionary      | key_key=master2 | key1=value21 | key2=value22 |          |
+        | ${dict3}    | Create Dictionary      | key_key=master3 | key1=value31 | key2=value32 |          |
+        | ${dict4}    | Create Dictionary      | key_key=master4 | key5=value41 | key6=value42 |          |
+        | ${the_list} | Create List            | ${dict1}        | ${dict2}     | ${dict3}     | ${dict4} |
+        | ${result}   | List Get Dict By Value | ${the_list}     | key_key      | master4      |          |
+
+        Variable ${result} has following structure:
+        | ${result} = {
+        |   'key_key': 'master4',
+        |   'key5': 'value41',
+        |   'key6': 'value42'
+        | }
+        """
+        try:
+            if match == 'first':
+                return input_list[next(index for (index, d) in enumerate(input_list) if d[key_name] == value)]
+            else:
+                return [d for d in input_list if d[key_name] == value]
+        except (KeyError, TypeError, StopIteration):
+            raise KeyError('list does not contain a dictionary with key:value "%s:%s"' % (key_name, value))
+
+    @staticmethod
+    def to_camel_case(snake_str, first_uppercase=False):
+        components = snake_str.split('_')
+        # We capitalize the first letter of each component except the first one
+        # with the 'title' method and join them together.
+        return (components[0] if not first_uppercase else components[0].title()) + ''.join(x.title() for x in components[1:])