AI Tools: Use Context Files (.cursorrules, AGENTS.md, CLAUDE.md) Without Bloat

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-tools-context-files-and-rules-r8a1-creators

1. Why is it important to keep project context files under 200 lines?

   1. Files over 200 lines automatically trigger security warnings in AI tools
   2. Shorter files are required by the file format specifications
   3. AI models can only process exactly 200 lines of context before forgetting the rest
   4. Shorter files cost less to process with every AI prompt and are more likely to be read

2. What does the lesson recommend doing with a rules file that exceeds 30 important rules?

   1. Cut to the 30 most important rules and move the rest to linked documentation
   2. Keep all rules in one file since AI models can process unlimited content
   3. Split rules into separate files for each programming language used
   4. Delete any rule older than six months regardless of importance

3. What is a fundamental limitation of context files?

   1. They can encourage AI behavior but cannot force the model to follow rules
   2. They cannot be used with version control systems
   3. They automatically enforce security policies without additional tools
   4. They replace the need for human code review entirely

4. Where should rules that must always be enforced (like security requirements) be placed?

   1. In a separate file that the AI reads first
   2. In the context file with a warning label
   3. In comments within the source code only
   4. In linting rules or CI pipelines

5. What should you do when two rules in your context file contradict each other?

   1. Keep both rules and let the AI decide which to follow
   2. Combine both rules into a single longer rule
   3. Flag the contradiction and resolve it before using the file
   4. Remove the newer rule to avoid conflicts

6. What is 'rule bloat' in the context of AI context files?

   1. When rules contain too many technical jargon terms
   2. When rules conflict with each other
   3. When context files grow excessively long, making them expensive and easy to ignore
   4. When too many developers contribute to the same file

7. What happens when a context file contradicts the diff (the specific changes being proposed)?

   1. The model is likely to ignore the context file in favor of the immediate code
   2. The AI automatically follows the diff because it's more recent
   3. The contradiction triggers an error in the AI tool
   4. The context file takes precedence over the diff

8. What is the primary purpose of citing paths to deeper documentation rather than inlining full content?

   1. To make the context file look more professional
   2. To ensure the AI cannot access the deeper documentation
   3. To satisfy file format requirements
   4. To reduce file size while keeping detailed information accessible

9. Why can context files not substitute for code review?

   1. Context files can only be read by AI tools, not humans
   2. The AI may ignore context file rules, so human judgment is still needed
   3. Context files are passive documents that cannot evaluate code changes
   4. AI models do not have access to the full codebase during review

10. What is meant by using context files for 'guidance' rather than 'gates'?

    1. Context files suggest behaviors but cannot enforce them; enforcement requires tools like lint
    2. Gates refer to authentication requirements for accessing the file
    3. Context files should only guide file organization, not code structure
    4. Guidance is for optional rules while gates are for required ones

11. Which statement best describes why context files are 'powerful'?

    1. They can execute code automatically
    2. They guarantee the AI will follow all listed rules
    3. They are seen by every AI prompt, giving consistent context
    4. They can replace database configuration

12. What type of information should lead (appear first) in a well-structured context file?

    1. General coding tips and best practices
    2. Non-negotiables like security requirements and coding conventions
    3. Examples of common mistakes to avoid
    4. Historical decisions and their justifications

13. What does it mean that context files 'train AI tools to ignore them'?

    1. Excessive rules cause the AI to treat all rules as suggestions rather than priorities
    2. Context files overwrite the AI's training data
    3. The AI becomes dependent on context files and cannot function without them
    4. AI tools learn to automatically delete long context files

14. Why should detailed technical documentation not be inlined in context files?

    1. AI tools cannot read detailed documentation
    2. Inlined documentation increases token costs on every call and dilutes important rules
    3. Inlined documentation causes parsing errors
    4. Technical documentation must be stored in databases

15. What happens if you put enforcement rules (like security policies) only in a context file without lint or CI?

    1. The rules may be ignored since context files only provide guidance
    2. The context file will prevent any code from being written that violates the rules
    3. The rules will be automatically enforced by the AI
    4. The AI will follow these rules more carefully than others