diff --git a/README.md b/README.md
index c6be72c14..d204f348d 100644
--- a/README.md
+++ b/README.md
@@ -64,21 +64,24 @@ The extraction, filtering and curation are done following the workflow below and
     As explained in the decision tree above, workflows are filtered first on EDAM terms (topics and operations), then on tags, workflow name and finally description based on the keywords provided in `keywords.yml` file.
 
 
-# Tools from RSEc
+## Extract tools from RSEC
 
-- Extract all metadata from RSEc/content/data 
+- Clone the RSEC repository and filter tools based on EDAM terms and keywords defined in `keywords.yml`. Validated tools are written to a JSON and a TSV file.
 
     ```
-    $ python bin/extract_rsec.py 
+    $ python bin/extract_rsec.py
     ```
 
-- Filter RSEc tools based on keywords and EDAM terms, place them in content/rsec
+    Custom output paths can be provided if needed:
 
     ```
-    $ python bin/filter_rsec.py
+    $ python bin/extract_rsec.py \
+        --kw keywords.yml \
+        --json-output content/rsec/infos/validated_tools_metadata.json \
+        --tsv-output content/rsec/infos/validated_tools_summary.tsv
     ```
-    As explained in the decision tree above, workflows are filtered first on EDAM terms (topics and operations), then on tags, workflow name and finally description based on the keywords provided in `keywords.yml` file. 
-    Workflows are filtered first on EDAM terms (topics and operations), then on tags, workflow name and finally description based on the keywords provided in "keywords.yml". 
+
+    Tools are filtered against three successive criteria: EDAM topics/operations from bio.tools metadata, keywords and acronyms from BioContainers metadata, then the same matching applied to free-text descriptions from bio.tools, BioContainers and Galaxy. A tool is kept if it passes any of them.
 
 # Bioconda