1. Colon (:)
Use: Introduces a list, label, or clarifies a prompt.
Example:
“Summarize the following patient’s condition: fever, high BP, and cough.”
2. Comma (,)
Use: Separates items in a list or clauses in a sentence.
Example:
“List the medications: aspirin, metformin, and lisinopril.”
3. Quotation Marks (” “)
Use: Indicate text or speech, define a specific string of input, or highlight a section.
Example:
“Summarize the patient’s report with the phrase ‘unremarkable physical exam’.”
4. Parentheses (( ))
Use: Add optional details, conditions, or clarification within a prompt.
Example:
“Provide details on the patient’s diagnosis (if available).”
5. Brackets ([ ])
Use: Include optional sections, variables, or alternatives.
Example:
“Summarize [recent labs] for this patient.”
6. Curly Braces ({ })
Use: Define a placeholder or variable that needs to be replaced with specific input.
Example:
“Provide a summary for {patient_name} diagnosed with {disease}.”
7. Pipe (|)
Use: Indicates a choice between multiple options.
Example:
“Generate a report based on the condition: asthma | diabetes.”
8. Hyphen (-)
Use: To create bullet points, lists, or connect two related items.
Example:
“- Patient history: Hypertension, Diabetes Mellitus.”
9. Asterisk (*)
Use: Highlight important terms or items, or indicate a wildcard.
Example:
“Important: The patient has a history of heart disease.”
10. Slash (/)
Use: Denote alternatives or combine related terms.
Example:
“Review patient’s diet/exercise plan.”
11. Equal Sign (=)
Use: Assign values, such as parameters, variables, or conditions.
Example:
“Patient’s age = 45.”
12. Semicolon (;)
Use: Separate related but distinct ideas or commands.
Example:
“List the following vitals; temperature, heart rate, and BP.”
13. Angle Brackets (< >)
Use: Indicate placeholders for content to be inserted.
Example:
“Write a summary for <patient_name>.”
14. Ampersand (&)
Use: Connect related elements or items.
Example:
“Evaluate heart rate & blood pressure.”
15. Ellipsis (…)
Use: Indicate continuation, truncation, or omitted content.
Example:
“Please provide details for the last lab results…”
16. Backslash ()
Use: Escape special characters or include literal interpretation of delimiters.
Example:
“Summarize with the phrase: “No significant findings”.”
17. At Symbol (@)
Use: Mention specific users or handle variable notation in some programming.
Example:
“Send results to @lab_technician.”
18. Keywords/Directives
Use: Instruct the AI to perform specific tasks such as summarizing, explaining, or generating content.
Example:
“Summarize the patient’s lab results and explain the potential treatment options for high blood sugar.”
19. Templates
Use: Predefined frameworks for consistency across repeated tasks.
Example:
“Use the following template to create a discharge summary:
Patient: {name}, Age: {age}, Diagnosis: {diagnosis}, Treatment: {treatment_plan}.”
20. Hierarchical Structure (Indented Lists)
Use: Organizing content into levels of importance or categories.
Example:
“Diagnosis:
Primary: Hypertension
Secondary: Diabetes.”
21. Formatting Modifiers (Bold, Italics, Underline)
Use: To emphasize key text by formatting.
Example:
“Explain hypertension in simple terms, focusing on treatment options.”
22. Conditional Logic (If-Then Statements)
Use: Instruct AI based on specific conditions.
Example:
“If the patient’s heart rate is above 100 bpm, recommend a cardiology consult.”
23. Enumerations
Use: Numbered or lettered lists to maintain order.
Example:
“List the top 3 treatment options for hypertension:
- Option 1
- Option 2
- Option 3.”
24. Questions (Interrogative Prompts)
Use: Frame prompts as questions to get specific information.
Example:
“What are the most common causes of high blood pressure?”
25. Time Modifiers
Use: Add time constraints or references to focus on specific timelines.
Example:
“Provide a treatment plan for diabetes over the next 6 months.”
26. Tone Modifiers
Use: Adjust the tone of the response (formal, casual, patient-friendly).
Example:
“Explain diabetes management in a patient-friendly tone.”
27. Prioritization Instructions
Use: Instruct AI to prioritize key information.
Example:
“List the side effects of metformin, prioritizing the most common.”
28. Constraints/Limiters
Use: Set boundaries on the length or type of response.
Example:
“Summarize the patient’s medical history in under 50 words.”
29. Examples or Sample Inputs
Use: Provide specific examples to guide the AI’s response.
Example:
“Write a peer review letter using this example: ‘The patient’s request for an MRI was denied due to…'”
30. File or Data Reference
Use: Refer to specific files or databases for the AI to base its response on.
Example:
“Analyze the patient’s lab results from {file_name} and suggest a treatment plan.”
31. New Line Indicator (\n)
Use: Directs AI to insert a line break or start a new line in the output.
Example:
“List the patient’s medications like this:
Aspirin 81 mg\nMetformin 500 mg\nLisinopril 10 mg.”
32. OR (Alternatives)
Use: Indicates options between two or more choices.
Example:
“Should the patient be treated with medication OR surgery?”
33. Forward Slash (//)
Use: Can be used for comments or separating different sections of a prompt (commonly used in programming or pseudo-code).
Example:
“Calculate the dosage for the patient // if weight exceeds 75 kg.”
34. Plus Sign (+)
Use: Adds items or combines multiple elements together.
Example:
“Summarize the combined effects of medication1 + medication2 on hypertension.”
35. Caret (^)
Use: Used for indicating power or importance, such as emphasizing dosage in a medical instruction.
Example:
“Make sure to adjust the dosage to ^{dosage_level} for elderly patients.”
36. Percent Sign (%)
Use: Represents a proportion or a percentage, useful for statistical data or lab results.
Example:
“List the patient’s cholesterol levels if they are above 70%.”
37. Underscore (_)
Use: Can be used for indicating a space or to emphasize a word or phrase in certain text processing systems.
Example:
“The patient’s vital signs show significant improvement.”
38. Backticks (`)
Use: Common in code or markdown formatting to enclose text for special formatting like monospaced fonts or commands.
Example:
“Write the command: python3 script.py to analyze the data.”
39. Tilde (~)
Use: Represents approximation or similarity, often used in text or scientific notations.
Example:
“Provide an approximation for the patient’s recovery time ~6 months.”
40. Double Hyphen (–)
Use: Can act as a delimiter or separator between sections in a prompt, or for indicating comments in coding prompts.
Example:
“List the patient’s symptoms — focus on primary diagnosis.”
41. Question Mark (?)
Use: Asks a question to get specific information or clarify something.
Example:
“What are the potential side effects of metformin?”
42. Conditional Formatting (Advanced If-Else Logic)
Use: Provides more detailed conditions and branching logic based on different criteria.
Example:
“If the patient’s BP > 140/90, suggest medication A. Else, recommend lifestyle changes.”
43. Switch Case Statements
Use: A way to handle multiple conditions by switching between cases.
Example:
“Switch based on patient age:
- Case <30: suggest aggressive treatment
- Case 30-60: recommend moderate intervention
- Case >60: suggest conservative approach.”
44. Looping Statements (For, While Loops)
Use: Loops can be used for repetitive tasks or iterating over data.
Example:
“List all the medications prescribed to the patient. For each medication, list its dosage and frequency.”
45. Variables Inside Variables (Nested Placeholders)
Use: Allows embedding placeholders inside other placeholders for dynamic content.
Example:
“Summarize the condition of {patient_name}, aged {patient_age}, diagnosed with {condition_1} and {condition_2}.”
46. Escaped Characters
Use: Handles characters that usually have special meaning in text (like quotes or symbols) and renders them as plain text.
Example:
“Explain the term “hypertension” and its impact on “heart disease”.”
47. Regular Expressions (RegEx)
Use: Advanced string-matching patterns, often used for pattern recognition or extracting data from complex strings.
Example:
“Find all numeric values in the string: {lab_results} using regular expressions.”
48. Flags or Parameters (Command-Line Style)
Use: Used to specify parameters, options, or settings within a prompt.
Example:
“Generate a report –include_labs –sort_by_date –output_format=pdf.”
49. Max/Min Functions
Use: Allows prompting for the maximum or minimum value in a set or list.
Example:
“Return the maximum BP reading from the patient’s history.”
50. Function Calls (Call-Based Prompts)
Use: Refers to invoking certain predefined actions or commands to generate responses.
Example:
“Call generate_diagnosis_report({patient_data}) to create the diagnosis summary.”
51. Nested Functions
Use: Combine multiple function calls in a single prompt to achieve complex results.
Example:
“First, generate a report using create_summary(), then analyze it with evaluate_summary().”
52. Scope Delimiters
Use: Control the scope or limit of a prompt, especially when multiple variables or elements are involved.
Example:
“Evaluate {test_results} within the scope of the last 3 months only.”
53. Data Retrieval and References
Use: Refers to a specific dataset or previously stored information.
Example:
“Use the dataset from {previous_lab_tests} to generate trends.”
54. Multiple Formatting Layers
Use: Apply multiple formatting rules (bold, italics, underlining) simultaneously to emphasize critical points.
Example:
“Ensure the patient avoids salt due to hypertension.“
55. Enumerating with References (Cross-Referencing)
Use: Create lists or numbers that refer back to earlier parts of the prompt for consistency.
Example:
“In section 1, discuss {condition}. Then in section 3, refer back to section 1’s explanation for more detail.”
56. Priority Assignment
Use: Assigns different weights or priorities to various actions or information.
Example:
“Assign a priority level 1 to all heart-related conditions, and level 2 to respiratory issues.”
57. Multi-Part Prompts
Use: Breaks down a prompt into multiple stages or parts to ensure more accurate results.
Example:
“Part 1: Summarize the patient’s condition.
Part 2: Suggest a treatment plan.
Part 3: Outline follow-up steps.”
58. Data Chaining (Chained Prompts)
Use: Sequentially link outputs of one prompt as the input of another.
Example:
“Use the output of the diagnosis report to generate the treatment plan in a follow-up prompt.”
59. Clustering or Grouping
Use: Group similar variables, conditions, or treatments to organize large datasets or lists.
Example:
“Group medications based on their target system: cardiovascular, respiratory, etc.”
60. Optimization and Constraints
Use: Solves problems while adhering to certain constraints (used in complex decision-making).
Example:
“Optimize the patient’s medication list while minimizing potential drug interactions.”
61. Weighted Decision Making
Use: Involves assigning weights to various choices and selecting the most optimal outcome.
Example:
“Choose the best treatment for the patient based on effectiveness (70%), side effects (20%), and cost (10%).”
62. Randomization or Shuffling
Use: Introduces an element of randomness or change in order within prompts.
Example:
“Shuffle the list of medications and recommend one at random for patient review.”
63. Memory and Contextual Awareness
Use: Incorporate memory to store and recall past information in future interactions (AI systems with memory).
Example:
“Remember the patient’s preferences from our previous conversation and apply them to this new treatment plan.”
