diff --git a/experimental/README.md b/experimental/README.md
new file mode 100644
index 000000000..528a79b20
--- /dev/null
+++ b/experimental/README.md
@@ -0,0 +1,73 @@
+# Prerequisites
+
+To compile this project you have to install the Rust language. You can find
+the installation instructions [here](https://www.rust-lang.org/tools/install).
+
+Once done, you can compile it by running:
+
+```bash
+cargo build
+```
+
+# Description
+
+## generic-snmp
+
+To run the project, you can use the following command:
+
+```bash
+cargo run -- -H localhost -v 2c -c public -j test.json
+```
+
+## conn library
+
+You have to install the Platypus library for Perl. You can do it by running:
+
+```bash
+cpanm FFI::Platypus::Lang::Rust
+```
+
+which installs the Perl bindings for Rust.
+
+Then you have to compile the project (as described in the previous section).
+
+Then, there is an example of Perl project using it in the perl directory you
+can start with:
+
+```
+cd perl
+./with-ffi.pl
+```
+
+# the generic-snmp program
+
+This is the main program of this directory. It is a generic SNMP client that
+can be used to query SNMP agents. Its goal is to be able to query any SNMP
+agent, no matter the MIBs it supports, you just have to fill a JSON file and
+it works.
+
+Its API is work in progress, but you can already use it to query SNMP agents.
+
+Here is an example of JSON file already supported:
+
+```
+{
+  "leaf": {
+    "name": "cpu",
+    "output": "{status}: {count} CPU(s) average usage is {total_cpu_avg} %",
+    "entries": [
+      { "Query": { "name": "cpu_{idx}", "oid": "1.3.6.1.2.1.25.3.3.1.2", "query": "Walk" }},
+      { "Agregation": { "name": "total_cpu_avg", "op": "Average"}}
+    ]
+  }
+}
+```json
+
+In this example, the output is built using several internal variables that are:
+* status: the status of the query (OK, WARNING, CRITICAL, UNKNOWN)
+* count: the number of variables returned by the query
+* total_cpu_avg: here, it is the average of the CPU usage, but it is defined in the Agregation entry with the name label.
+
+So you could also define others variables for the output.
+
+In the query entry, there is also an `idx` variable that is an integer enumerating the entries og the query. It is used to build the name of the variable in the output. It starts from 0.