diff --git a/spec.md b/spec.md
index 7d2245b..163d2cf 100644
--- a/spec.md
+++ b/spec.md
@@ -4,25 +4,54 @@
 and the content of wikimedia pages. This library must ne built on top of the 2 existing rust libraries
 parse_wiki_text and parse_wiki_text to implement those functionalities
 
-**User Story 1**: As a user, I want to get the raw text content of a wikimedia article by specifying the site, language and article name.
+**User Story 1**: As a user, I want to get the raw text content of a Wikimedia article by specifying the site, language and article name.
 **User Story 2**: As a user, I want to get the list of all the links of an article
 **User Story 3**: As a user, I want to get the list of all the categories of an article
 **User Story 4**: As a user, I want to get the list of all the references of an article
 **User Story 5**: As a user, I want to get the list of all the templates of an article
 **User Story 6**: As a user, I want to get the list of all the parameters of a template
+**User Story 7**: As a user, I want to get the list of all the sections of an article
 
 **Functional Requirements**:
-1. Use mediawiki_rest_api to get access to the wikimedia article
+1. Use mediawiki_rest_api to get access to the Wikimedia article
 2. Use mediawiki_rest_api to get the initial content of the article
-3. Use parse_wiki_text to parse the wikimedia article
-4. Use parse_wiki_text to interact with any internal content of the article
-5. Use parse_wiki_text nodes
-6. Always write unit tests for every functions
+3. Use parse_wiki_text to parse the Wikimedia article
+4. Use parse_wiki_text to interact with any internal content of the article 
+5. Use parse_wiki_text nodes as a main way to manage data access 
+6. The central data structure is a WikiPage with its associated trait of the same name. This structure has the following fields:
+   - language -> Language of the page
+   - title -> Title of the page
+   - content -> raw text content of the article, output of the MediaWikiPage get function
+   - parsed -> node based structure, output of the Configuration parse function
+7. Organize the functions to be implementation as a Trait 
+8. Only language and article name are known at initialization 
+9. All other functions including fetching and parsing the page are performed after initial creation of the structure
+10. Some specific low level functions must created to help building higher level functionalities. These functions are:
+    - get_links
+    - get_categories
+    - get_references
+    - get_templates
+    - get_template_parameters
+    - get_sections
+11. When interacting with a page the steps are:
+    - Create a new WikiPage with the language and title
+    - Load the content and store the results in the content field
+    - Parse the content and store the results in the parsed field
+12. After those initializations steps, all functions should interact with the fields directly and do not reload the values.
+
+**Critical Rules**:
+
+**Testing Requirements**:
+1. Always write unit tests for every functions
+2. Write all the tests in the dedicated /tests folder. Do not include any test in the main code folders and files
+3. 
 
 **Technical Constraints**:
 - rust language
 - mediawiki_rest_api
 - parse_wiki_text
+- use the GEMINI.md file for specific code rules
+- 
 
 **Test Cases**:
 - **Scenario A**: Valid access -> Status 200 + content of the article.
@@ -31,4 +60,5 @@ parse_wiki_text and parse_wiki_text to implement those functionalities
 - **Scenario D**: Valid access -> Status 200 + references of the article.
 - **Scenario E**: Valid access -> Status 200 + templates of the article.
 - **Scenario F**: Valid access -> Status 200 + parameters of a template.
+- **Scenario G**: Valid access -> Status 200 + sections of the article.
 - **Scenario Z**: Incorrect article -> Status 401 + error message.