diff --git a/steps/README.md b/steps/README.md
new file mode 100644
index 00000000..d4625b28
--- /dev/null
+++ b/steps/README.md
@@ -0,0 +1,20 @@
+## Gathering Steps
+
+The entire process of gathering metrics from GitHub repos consists of these
+steps:
+
+* **Discovering**. Here we fetch the list of repos from GitHub and then create
+directories for them.
+* **Polishing**. Then we delete directories that don't exist in the list of
+required repositories.
+* **Unregistering**. During this step, we clean directories from the CSV
+register if their clones are absent.
+* **Cloning**. In this step we run `git clone` on found repositories.
+* **JPEEK**. Here, we build those gathered repositories and run
+[jpeek](https://github.com/cqfn/jpeek) on them.
+* **Filtering**. Is where we apply [all the filters](https://github.com/yegor256/cam/tree/master/filters)
+in order to get rid of irrelevant classes (such as `*Test`, `*ITCase`, invalid
+files and so on). Whole filtering process will be printed in the final report,
+you can check it [here](http://cam.yegor256.com/cam-2024-03-02.pdf).
+* **Measuring**. We calculate metrics for each file using these [metrics](https://github.com/yegor256/cam/tree/master/metrics).
+* **Aggregating**. Finally, we aggregate all metrics in summary CSV files.
diff --git a/steps/discover-repos.rb b/steps/discover-repos.rb
index acae423f..28cc9c39 100755
--- a/steps/discover-repos.rb
+++ b/steps/discover-repos.rb
@@ -88,7 +88,8 @@
       created_at: i[:created_at].iso8601,
       size: i[:size],
       open_issues_count: i[:open_issues_count],
-      description: i[:description]
+      description: i[:description],
+      topics: i[:topics]
     }
     puts "Found #{i[:full_name].inspect} GitHub repo ##{found.count} \
 (#{i[:forks_count]} forks, #{i[:stargazers_count]} stars)"
diff --git a/tests/steps/test-discover-repos.sh b/tests/steps/test-discover-repos.sh
index c26850fd..0df2891b 100755
--- a/tests/steps/test-discover-repos.sh
+++ b/tests/steps/test-discover-repos.sh
@@ -37,7 +37,7 @@ tex=${TARGET}/foo.tex
     test -e "${csv}"
     test -s "${tex}"
     test "$(wc -l < "${csv}" | xargs)" = '4'
-    test "$(head -1 "${csv}" | tr "," "\n" | wc -l | xargs)" = '8'
+    test "$(head -1 "${csv}" | tr "," "\n" | wc -l | xargs)" = '9'
 } > "${stdout}" 2>&1
 echo "👍🏻 Small repositories discovery test is succeed"
 
@@ -50,6 +50,6 @@ echo "👍🏻 Small repositories discovery test is succeed"
     test -e "${csv}"
     test -s "${tex}"
     test "$(wc -l < "${csv}" | xargs)" = '36'
-    test "$(head -1 "${csv}" | tr "," "\n" | wc -l | xargs)" = '8'
+    test "$(head -1 "${csv}" | tr "," "\n" | wc -l | xargs)" = '9'
 } > "${stdout}" 2>&1
 echo "👍🏻 Medium repositories discovery test is succeed"