is https://user:pass@foo.com a webhook? yes
is blah://foo/bar a webhook? no
UNMASK RESULT This is a test ~FN:hangup()-; 1234 ~FN:hangup()-;
STRIP UNMASK RESULT (null)
Number [+1 213-555-1212] formatted to [(213) 555-1212]
TOKENS EN 374
TOKENS FG 374
TOKENS FR 400
TTS NUMBER [Testing numbers  + 1,. 4 1 4,. 5 5 5,. 1 2 1 2 and + 1,. 2 1 2,. 4 5 5,. 1 2 1 2 and  + 1 2 1 2 1 2 3 1 2 3 4,. and  + 1,. 3 3 2,. 3 4 5,. 1 2 3 4 and  + 4 4,. 1 1 7,. 2 3 4,. 5 6 7 8 and + 1,. 9 0 8,. 4 3 4,. 9 5 7 8 ]
TTS NUMBER [I heard your phone number as  2 6 2,. 2 5 2,. 3 2 6 2 —is that correct?]
REPLACE TEST what is SWML?
R: [bar]
R: [{
	"number":	55
}]
R: [29]
R: [55]
NULL
R: [12.500000]
R: [ex1 LONGER stRing]
R: [cval2]
R: [phone national 2___foobar___6___foobar___2___foobar___3___foobar___3___foobar___4___foobar___5___foobar___4___foobar___0___foobar___0]
R: [phone international +1 262-334-5400]
R: [phone RFC3966 tel:+1-262-334-5400]
R: [phone e164 +12623345400]
R: [NESTED 'a' 'ex1 LONGER stRing']
R: [EMPTY 'a' '']
R: [function 'bar' 'ex1 LONGER stRing' and ]
R: [expr function 2+2=4]
R: [tz function  06/09/25 11:49:18]
R: [the answer is 'bar' and '12.500000' and {ex1%20longer%20string} the end]
R: [colon test ]
R: [the undef vars answer is  and  and () the end]
R: [no vars in this longer ${ string what will happen?]
R: [end on half var ${]
R: [end on null var ]
R: [empty var in this longer  string what will happen?]
R: [empty var in this longer  string what will happen?]
R: [lots o' dots  string what will happen?]
R: [url encode the val of grr%3Dthis%20and%20that ok?]
R: [array test url encode the val of Orlando ok?]
R: [array test url encode the val of Milwaukee,%20WI ok?]
R: [array test url encode the val of  ok?]
OLD {
	"name":	"fred\"\\",
	"age":	25,
	"string":	"${name} ${age}"
}
NEW {
	"name":	"fred\"\\",
	"age":	25,
	"string":	"fred\"\\ 25"
} 0x6060024c46a0The string contains an emoji.
stripped string [Hello, world! Olá]
COUNT 1
DONE
[OK, goodbye ~FN:assistant_hangup()-;][OK, goodbye **assistant_hangup**]
PARSE FIRST [ln][English]


------
ORIG: [~LN(English)-; Greetings, I am Franklin, an expert at all things Star Wars. How can I assist you today? ~LN(English)-; Greetings, I am Franklin, an expert at all things Star Wars. How can I assist you today?
]
STRIPPED 1, x: [ Greetings, I am Franklin, an expert at all things Star Wars. How can I assist you today?  Greetings, I am Franklin, an expert at all things Star Wars. How can I assist you today?
]
PREFIXES 0, x: [(null)] (nil)
SKIP [Greetings, I am Franklin, an expert at all things Star Wars. How can I assist you today? ~LN(English)-; Greetings, I am Franklin, an expert at all things Star Wars. How can I assist you today?
]
SPACE TEST [1][0][0][1][1]
Orig [https://me:mine@cool.com/test.cgi?1234] Ret [1] Parsed [https://cool.com/test.cgi?1234] User [me] Pass [mine]
Orig [https://user:pass@cool.com/test.cgi?1234] Ret [1] Parsed [https://cool.com/test.cgi?1234] User [user] Pass [pass]
Orig [https://cool.com/test.cgi?1234] Ret [0] Parsed [(null)] User [(null)] Pass [(null)]
Orig [https://user:pass@subdomain.cool.com/path] Ret [1] Parsed [https://subdomain.cool.com/path] User [user] Pass [pass]
Orig [https://user@domain.com:1234@myhandler.com/myscript] Ret [1] Parsed [https://myhandler.com/myscript] User [user@domain.com] Pass [1234]
Orig [https://user:#$%@@cool.com/test.cgi?1234] Ret [1] Parsed [https://cool.com/test.cgi?1234] User [user] Pass [#$%@]
Orig [https://user:%23$%25%40@cool.com/test.cgi?1234] Ret [1] Parsed [https://cool.com/test.cgi?1234] User [user] Pass [#$%@]
Orig [https://user::::@@@@cool.com/test.cgi?1234] Ret [1] Parsed [https://cool.com/test.cgi?1234] User [user] Pass [:::@@@]
Orig [https://user:pass with space@cool.com/test.cgi] Ret [1] Parsed [https://cool.com/test.cgi] User [user] Pass [pass with space]
Orig [https://::::@domain.com/foo.cgi] Ret [0] Parsed [(null)] User [(null)] Pass [(null)]
Orig [user@@@@@@domain.com] Ret [0] Parsed [(null)] User [(null)] Pass [(null)]
Orig [https://user@@@@@@domain.com/foo] Ret [1] Parsed [https://@@@@@domain.com/foo] User [user] Pass []
Orig [https://::::::::::::@domain.com/path] Ret [0] Parsed [(null)] User [(null)] Pass [(null)]
Orig [https://@domain.com/path] Ret [0] Parsed [(null)] User [(null)] Pass [(null)]
Orig [https://:@domain.com/path] Ret [0] Parsed [(null)] User [(null)] Pass [(null)]
Orig [https://user:@domain.com/path] Ret [1] Parsed [https://domain.com/path] User [user] Pass []
Orig [https://user:pass@co%6Fl.com/test] Ret [1] Parsed [https://co%6Fl.com/test] User [user] Pass [pass]
Orig [https://user%20name:pass@domain.com/path] Ret [1] Parsed [https://domain.com/path] User [user name] Pass [pass]
Orig [https://user:pa%00ss@domain.com/path] Ret [1] Parsed [https://domain.com/path] User [user] Pass []
Orig [https://user@dom%41in.com/path] Ret [1] Parsed [https://dom%41in.com/path] User [user] Pass []
Orig [https://usernamewithaverylongnamethatexceedstypicallengthlimits:passwordwithaverylongnametotestlimits@cool.com/path] Ret [1] Parsed [https://cool.com/path] User [usernamewithaverylongnamethatexceedstypicallengthlimits] Pass [passwordwithaverylongnametotestlimits]
Orig [https://user::::pass::::@domain.com/path] Ret [1] Parsed [https://domain.com/path] User [user] Pass [:::pass::::]
Orig [https://user:user:user:user:pass@domain.com/path] Ret [1] Parsed [https://domain.com/path] User [user] Pass [user:user:user:pass]
Orig [user:pass@domain.com] Ret [0] Parsed [(null)] User [(null)] Pass [(null)]
Orig [https://user:pass@] Ret [0] Parsed [(null)] User [(null)] Pass [(null)]
Orig [https:///path] Ret [0] Parsed [(null)] User [(null)] Pass [(null)]
Orig [https://user:pass@:/path] Ret [0] Parsed [(null)] User [(null)] Pass [(null)]
J: {
	"parsed":	[{
			"data":	[{
					"question":	"What is your name?",
					"answer":	"Tony"
				}, {
					"question":	"What is your job?",
					"answer":	"CEO"
				}, {
					"question":	"Have you ever used SignalWire products before?",
					"answer":	"Yes"
				}, {
					"question":	"If yes, Do you have any feedback?",
					"answer":	"Better documentation"
				}]
		}],
	"raw":	"This is json. {\"data\": [{\"question\": \"What is your name?\", \"answer\": \"Tony\"}, {\"question\": \"What is your job?\", \"answer\": \"CEO\"}, {\"question\": \"Have you ever used SignalWire products before?\", \"answer\": \"Yes\"}, {\"question\": \"If yes, Do you have any feedback?\", \"answer\": \"Better documentation\"}]}",
	"substituted":	"This is json. "
}
Input: '1h' => 1 seconds
Input: '30m' => 30 seconds
Input: '45s' => 45 seconds
Input: '1:30' => 90 seconds
Input: '00:30' => 30 seconds
Input: '1:2:30' => 3750 seconds
Input: '1234567890' => 1234567890 seconds
Input: 'invalid' => Error: Invalid format
Input: '-1h' => Error: Invalid characters in input
Input: '1:::30' => Error: Consecutive colons detected
Input: '999999999999999999999h' => Error: Numeric overflow
Input: '   1h   ' => 1 seconds
Input: '1h30m' => 130 seconds
Input: '1m-30s' => Error: Invalid characters in input
Input: '1:30:60' => Error: Invalid format
Input: '1:2:3:4' => Error: Too many colons in input
Input: '12345678901234567890h' => Error: Numeric overflow
Input: '@#$%' => Error: Invalid characters in input
Test case 0:
Original:
No markdown here.

Stripped:
No markdown here.

Test case 1:
Original:
# Header Text
Some content.

Stripped:
Header Text
Some content.

Test case 2:
Original:
#include <stdio.h>

Stripped:
#include <stdio.h>

Test case 3:
Original:
A paragraph with *italic* and **bold** text.

Stripped:
A paragraph with italic and bold text.

Test case 4:
Original:
> A blockquote line

Stripped:
A blockquote line

Test case 5:
Original:
- A bullet list item

Stripped:
A bullet list item

Test case 6:
Original:
1. An ordered list item

Stripped:
1. An ordered list item

Test case 7:
Original:
```c
int main() {
    return 0;
}
```

Stripped:

int main() {
    return 0;
}

Test case 8:
Original:
Inline code: `printf("Hello, World!");` remains.

Stripped:
Inline code: printf("Hello, World!"); remains.

Test case 9:
Original:
[Link text](http://example.com)

Stripped:
Link text

Test case 10:
Original:
![Image alt](http://example.com/image.png)

Stripped:
Image alt

Test case 11:
Original:
Mixed [link](http://example.com) and `code` in one line.

Stripped:
Mixed link and code in one line.

Test case 12:
Original:
#include <stdio.h>

int main() {
    printf("Hello, World!\n");
    return 0;
}
]

Stripped:
#include <stdio.h>

int main() {
printf("Hello, World!\n");
return 0;
}
]

Test case 13:
Original:
**Sonic the Hedgehog 3** - This action-packed adventure is set to release on December nineteenth,

Stripped:
Sonic the Hedgehog 3 - This action-packed adventure is set to release on December nineteenth,

Test case 14:
Original:
## Subheader

This is a multi-line paragraph. It should maintain its spacing and
line breaks even after markdown is stripped.

Stripped:
Subheader

This is a multi-line paragraph. It should maintain its spacing and
line breaks even after markdown is stripped.

Test case 15:
Original:
> Blockquote with multiple lines
> that should be preserved in some form.

And this is a normal paragraph.

Stripped:
Blockquote with multiple lines
that should be preserved in some form.

And this is a normal paragraph.

Test case 16:
Original:
- List item one
- List item two
  - Nested item
- List item three

Stripped:
List item one
List item two
Nested item
List item three

Test case 17:
Original:
1. First ordered item
2. Second ordered item
3. Third ordered item

Stripped:
1. First ordered item
2. Second ordered item
3. Third ordered item

Test case 18:
Original:
A paragraph with multiple inline code elements: `code1` and then `code2` followed by text.

Stripped:
A paragraph with multiple inline code elements: code1 and then code2 followed by text.

Test case 19:
Original:
A paragraph with *emphasis*, **strong emphasis**, and ***both***.
Also, underscores _and_ tildes ~are~ present.

Stripped:
A paragraph with emphasis, strong emphasis, and both.
Also, underscores and tildes are present.

Test case 20:
Original:
Fenced code block with language specified:
```python
def hello_world():
    print('Hello, world!')
```
Another paragraph after the code block.

Stripped:
Fenced code block with language specified:

def hello_world():
    print('Hello, world!')

Another paragraph after the code block.

Test case 21:
Original:
A complex nested example:
# Header

Some introductory text.

> A blockquote with a list:
> - Item one
> - Item two

More text after the blockquote.

Stripped:
A complex nested example:
Header

Some introductory text.

A blockquote with a list:
Item one
Item two

More text after the blockquote.

Running XML escape tests
========================

Test 1:
Input: Hello world
Type: Plain text or text with special characters
Output: (NULL) - No escaping needed or NULL input

Test 2:
Input: Simple text without any special characters
Type: Plain text or text with special characters
Output: (NULL) - No escaping needed or NULL input

Test 3:
Input: Text with numbers 123 and symbols !@#$%^*()_+
Type: Plain text or text with special characters
Output: (NULL) - No escaping needed or NULL input

Test 4:
Input: This & that
Type: Plain text or text with special characters
Output: This &amp; that

Test 5:
Input: This < that
Type: Plain text or text with special characters
Output: This &lt; that

Test 6:
Input: This > that
Type: Plain text or text with special characters
Output: This &gt; that

Test 7:
Input: This " double quote
Type: Plain text or text with special characters
Output: This &quot; double quote

Test 8:
Input: This ' single quote
Type: Plain text or text with special characters
Output: This &apos; single quote

Test 9:
Input: Multiple & < > " ' characters in one string
Type: Contains XML tags but no <speak> tags
Output: Multiple &amp; &lt; &gt; &quot; &apos; characters in one string

Test 10:
Input: AT&T company name with ampersand
Type: Plain text or text with special characters
Output: AT&amp;T company name with ampersand

Test 11:
Input: 5 < 10 > 3 mathematical expressions
Type: Contains XML tags but no <speak> tags
Output: 5 &lt; 10 &gt; 3 mathematical expressions

Test 12:
Input: <emphasis>Emphasized text</emphasis>
Type: Contains XML tags but no <speak> tags
Output: <emphasis>Emphasized text</emphasis>

Test 13:
Input: <prosody rate="slow">Slow speech</prosody>
Type: Contains XML tags but no <speak> tags
Output: <prosody rate="slow">Slow speech</prosody>

Test 14:
Input: <break time="500ms"/>
Type: Contains XML tags but no <speak> tags
Output: <break time="500ms"/>

Test 15:
Input: <phoneme alphabet="ipa" ph="təˈmeɪtoʊ">tomato</phoneme>
Type: Contains XML tags but no <speak> tags
Output: <phoneme alphabet="ipa" ph="təˈmeɪtoʊ">tomato</phoneme>

Test 16:
Input: <s>This is a sentence.</s> <s>This is another.</s>
Type: Contains XML tags but no <speak> tags
Output: <s>This is a sentence.</s> <s>This is another.</s>

Test 17:
Input: <p>This is a paragraph with <b>bold</b> text.</p>
Type: Contains XML tags but no <speak> tags
Output: <p>This is a paragraph with <b>bold</b> text.</p>

Test 18:
Input: <say-as interpret-as="characters">SSML</say-as>
Type: Contains XML tags but no <speak> tags
Output: <say-as interpret-as="characters">SSML</say-as>

Test 19:
Input: <speak>Hello world</speak>
Type: Contains <speak> tags
Output: Hello world

Test 20:
Input: <speak>Simple speak tag</speak>
Type: Contains <speak> tags
Output: Simple speak tag

Test 21:
Input: <speak>Say <emphasis>this</emphasis> word</speak>
Type: Contains <speak> tags
Output: Say  word

Test 22:
Input: <speak>Count <say-as interpret-as="cardinal">123</say-as> now</speak>
Type: Contains <speak> tags
Output: Count  now

Test 23:
Input: <speak>This & that <break time="300ms"/> now</speak>
Type: Contains <speak> tags
Output: This &amp; that  now

Test 24:
Input: <speak>This <phoneme alphabet="ipa" ph="bɑ.skɛt">basket</phoneme> pronunciation</speak>
Type: Contains <speak> tags
Output: This  pronunciation

Test 25:
Input: <speak>Multiple <prosody rate="slow">nested <emphasis level="strong">tags</emphasis> with</prosody> special & characters</speak>
Type: Contains <speak> tags
Output: Multiple  special &amp; characters

Test 26:
Input: Text before <emphasis>emphasized</emphasis> and after
Type: Contains XML tags but no <speak> tags
Output: Text before <emphasis>emphasized</emphasis> and after

Test 27:
Input: Mix of text and <break time="200ms"/> pauses
Type: Contains XML tags but no <speak> tags
Output: Mix of text and <break time="200ms"/> pauses

Test 28:
Input: A sentence with <prosody pitch="high">higher pitch</prosody> in the middle
Type: Contains XML tags but no <speak> tags
Output: A sentence with <prosody pitch="high">higher pitch</prosody> in the middle

Test 29:
Input: Special characters & in <say-as interpret-as="spell-out">SSML</say-as> content
Type: Contains XML tags but no <speak> tags
Output: Special characters &amp; in <say-as interpret-as="spell-out">SSML</say-as> content

Test 30:
Input: Text with <phoneme alphabet="ipa" ph="ˈfiːnɪks">phoenix</phoneme> pronunciation and & ampersand
Type: Contains XML tags but no <speak> tags
Output: Text with <phoneme alphabet="ipa" ph="ˈfiːnɪks">phoenix</phoneme> pronunciation and &amp; ampersand

Test 31:
Input: Text before <speak>inside speak</speak> and after
Type: Contains <speak> tags
Output: Text before <speak>inside speak</speak> and after

Test 32:
Input: <speak>Outer <emphasis>inner</emphasis> tags</speak> and text after
Type: Contains <speak> tags
Output: Outer  tags

Test 33:
Input: Text with & special <speak>character and <break/> tag</speak>
Type: Contains <speak> tags
Output: Text with &amp; special <speak>character and <break/> tag</speak>

Test 34:
Input: Multiple <speak>speak tag 1</speak> and <speak>speak tag 2</speak> instances
Type: Contains <speak> tags
Output: Multiple <speak>speak tag 1</speak> and <speak>speak tag 2</speak> instances

Test 35:
Input: <speak>Normal <![CDATA[This & that < > ' " shouldn't be escaped]]> text</speak>
Type: Contains <speak> tags
Output: Normal This &amp; that < > &apos; &quot; shouldn&apos;t be escaped text

Test 36:
Input: <emphasis><![CDATA[CDATA without speak tags & < > ' "]]></emphasis>
Type: Contains XML tags but no <speak> tags
Output: <emphasis><![CDATA[CDATA without speak tags & < > ' "]]></emphasis>

Test 37:
Input: <![CDATA[Just CDATA with no tags & < > ' "]]>
Type: Contains XML tags but no <speak> tags
Output: &lt;![CDATA[Just CDATA with no tags &amp; &lt; &gt; &apos; &quot;]]&gt;

Test 38:
Input: <speak>Text <!-- Comment inside speak --> more text</speak>
Type: Contains <speak> tags
Output: Text  more text

Test 39:
Input: <!-- Comment outside speak --> Text <emphasis>tag</emphasis>
Type: Contains XML tags but no <speak> tags
Output: <!-- Comment outside speak --> Text <emphasis>tag</emphasis>

Test 40:
Input: Text with & character <!-- Comment with & < > ' " --> and after
Type: Contains XML tags but no <speak> tags
Output: Text with &amp; character &lt;!-- Comment with &amp; &lt; &gt; &apos; &quot; --&gt; and after

Test 41:
Input: <speak><?xml-custom data="value"?> Text inside</speak>
Type: Contains <speak> tags
Output:  Text inside

Test 42:
Input: <?xml version="1.0"?><emphasis>No speak tags</emphasis>
Type: Contains XML tags but no <speak> tags
Output: <?xml version="1.0"?><emphasis>No speak tags</emphasis>

Test 43:
Input: Text &amp; <?custom-pi?> processing instruction
Type: Contains XML tags but no <speak> tags
Output: Text &amp;amp; &lt;?custom-pi?&gt; processing instruction

Test 44:
Input: <speak><prosody rate='120%'>Fast & clear</prosody></speak>
Type: Contains <speak> tags
Output: <prosody rate="120%">Fast &amp;amp; clear</prosody>


Test 45:
Input: <prosody rate='120%'>Fast & clear without speak</prosody>
Type: Contains XML tags but no <speak> tags
Output: <prosody rate='120%'>Fast &amp; clear without speak</prosody>

Test 46:
Input: <speak attr="value&another">Invalid attribute</speak>
Type: Contains XML tags but no <speak> tags
Output: Invalid attribute

Test 47:
Input: Attribute with special chars <tag attr="value&another">text</tag>
Type: Contains XML tags but no <speak> tags
Output: Attribute with special chars &lt;tag attr=&quot;value&amp;another&quot;&gt;text&lt;/tag&gt;

Test 48:
Input: <speak>Unclosed tag <emphasis>will this work?</speak>
Type: Contains <speak> tags
Output: Unclosed tag 

Test 49:
Input: <emphasis>Unclosed tag without speak
Type: Contains XML tags but no <speak> tags
Output: <emphasis>Unclosed tag without speak

Test 50:
Input: Overlapping <a><b>tags</a></b>
Type: Contains XML tags but no <speak> tags
Output: Overlapping &lt;a&gt;&lt;b&gt;tags&lt;/a&gt;&lt;/b&gt;

Test 51:
Input: <a><b><c></b></a></c>
Type: Contains XML tags but no <speak> tags
Output: &lt;a&gt;&lt;b&gt;&lt;c&gt;&lt;/b&gt;&lt;/a&gt;&lt;/c&gt;

Test 52:
Input: Incorrectly & escaped
Type: Plain text or text with special characters
Output: Incorrectly &amp; escaped

Test 53:
Input: 
Type: Plain text or text with special characters
Output: (NULL) - No escaping needed or NULL input

Test 54:
Input: (NULL)
Type: Plain text or text with special characters
Output: (NULL) - No escaping needed or NULL input

Test 55:
Input:  
Type: Plain text or text with special characters
Output: (NULL) - No escaping needed or NULL input

Test 56:
Input: ><&"'
Type: Contains XML tags but no <speak> tags
Output: &gt;&lt;&amp;&quot;&apos;

Test 57:
Input: <>
Type: Contains XML tags but no <speak> tags
Output: &lt;&gt;

Test 58:
Input: <speak></speak>
Type: Contains <speak> tags
Output: 

Test 59:
Input: <speak>   </speak>
Type: Contains <speak> tags
Output:    

Test 60:
Input: <emphasis></emphasis>
Type: Contains XML tags but no <speak> tags
Output: <emphasis></emphasis>

Test 61:
Input: <speak><<>>&&''""</speak>
Type: Contains <speak> tags
Output: 

Test 62:
Input: <<>>&&''""
Type: Contains XML tags but no <speak> tags
Output: &lt;&lt;&gt;&gt;&amp;&amp;&apos;&apos;&quot;&quot;

Test 63:
Input: <speak><emphasis><prosody><say-as><break/></say-as></prosody></emphasis></speak>
Type: Contains <speak> tags
Output: <emphasis>
  <prosody>
    <say-as>
      <break></break>
    </say-as>
  </prosody>
</emphasis>


Test 64:
Input: <speak>Very <break time="100ms"/> very <break time="100ms"/> very & long <emphasis>string</emphasis> with <prosody rate="slow">many</prosody> tags</speak>
Type: Contains <speak> tags
Output: Very  very  very &amp; long  with  tags

Test 65:
Input: Very <break time="100ms"/> very <break time="100ms"/> very & long <emphasis>string</emphasis> with <prosody rate="slow">many</prosody> tags
Type: Contains XML tags but no <speak> tags
Output: Very <break time="100ms"/> very <break time="100ms"/> very &amp; long <emphasis>string</emphasis> with <prosody rate="slow">many</prosody> tags

Test 66:
Input: <div><speak>SSML inside HTML</speak></div>
Type: Contains <speak> tags
Output: <div><speak>SSML inside HTML</speak></div>

Test 67:
Input: <html><body>Text <speak>SSML</speak> more text</body></html>
Type: Contains <speak> tags
Output: <html><body>Text <speak>SSML</speak> more text</body></html>

Test 68:
Input: <speak>SSML <div>HTML inside SSML</div> more SSML</speak>
Type: Contains <speak> tags
Output: SSML  more SSML

Test 69:
Input: Ah, your name is <say-as interpret-as="characters"><prosody rate="slow">J U S T I N</prosody></say-as> <say-as interpret-as="characters">M A S S E Y</say-as>! Nice to meet you!
Type: Contains XML tags but no <speak> tags
Output: Ah, your name is <say-as interpret-as="characters"><prosody rate="slow">J U S T I N</prosody></say-as> <say-as interpret-as="characters">M A S S E Y</say-as>! Nice to meet you!

Rendered Markdown:


## Rules

High-level operational guidelines.

- Always respond.
- Keep tone consistent.

### Formatting

Use markdown syntax where appropriate.

- Use `**` for bold.
- Prefer lists over long paragraphs.

#### Advanced Markdown

Support nested lists and code blocks.

- Support triple-backtick blocks.
- Handle indentation properly.
Rendered XML:

<prompt>
  <section>
    <title>Rules</title>
    <body>High-level operational guidelines.</body>
    <bullets>
      <bullet>Always respond.</bullet>
      <bullet>Keep tone consistent.</bullet>
    </bullets>
    <subsections>
      <section>
        <title>Formatting</title>
        <body>Use markdown syntax where appropriate.</body>
        <bullets>
          <bullet>Use `**` for bold.</bullet>
          <bullet>Prefer lists over long paragraphs.</bullet>
        </bullets>
        <subsections>
          <section>
            <title>Advanced Markdown</title>
            <body>Support nested lists and code blocks.</body>
            <bullets>
              <bullet>Support triple-backtick blocks.</bullet>
              <bullet>Handle indentation properly.</bullet>
            </bullets>
          </section>
        </subsections>
      </section>
    </subsections>
  </section>
</prompt>




=======
JSON:
[{
		"title":	"Assistant Description",
		"body":	"You are a helpful AI Assistant named Franklin.",
		"numbered":	true
	}, {
		"title":	"Key Skills",
		"body":	"You have the following skills",
		"numberedBullets":	true,
		"bullets":	["You like starwars.", "You can look up the weather at any location (including starwars planets).", "You can dial digits when asked."],
		"subsections":	[{
				"numbered":	false,
				"title":	"Communication Skills",
				"body":	"You can explain things clearly."
			}, {
				"title":	"Utility Functions",
				"body":	"You can interact with external services."
			}],
		"numbered":	false
	}, {
		"title":	"At Conversation Start",
		"body":	"Introduce yourself and see if the user has any questions."
	}]
Rendered Markdown:


## 1. Assistant Description

You are a helpful AI Assistant named Franklin.

## 2. Key Skills

You have the following skills

1. You like starwars.
2. You can look up the weather at any location (including starwars planets).
3. You can dial digits when asked.

### Communication Skills

You can explain things clearly.

### Utility Functions

You can interact with external services.

## 3. At Conversation Start

Introduce yourself and see if the user has any questions.
Rendered XML:

<prompt>
  <section>
    <title>1. Assistant Description</title>
    <body>You are a helpful AI Assistant named Franklin.</body>
  </section>
  <section>
    <title>2. Key Skills</title>
    <body>You have the following skills</body>
    <bullets>
      <bullet index="1">You like starwars.</bullet>
      <bullet index="2">You can look up the weather at any location (including starwars planets).</bullet>
      <bullet index="3">You can dial digits when asked.</bullet>
    </bullets>
    <subsections>
      <section>
        <title>Communication Skills</title>
        <body>You can explain things clearly.</body>
      </section>
      <section>
        <title>Utility Functions</title>
        <body>You can interact with external services.</body>
      </section>
    </subsections>
  </section>
  <section>
    <title>3. At Conversation Start</title>
    <body>Introduce yourself and see if the user has any questions.</body>
  </section>
</prompt>




=======
JSON:
[{
		"title":	"System Persona",
		"body":	"Act as a security-focused backend engineer experienced with WebSockets and TLS."
	}, {
		"title":	"Problem Context",
		"body":	"A developer is trying to set up a secure WebSocket (wss://) server but is unsure about certificate handling and client validation."
	}, {
		"title":	"Response Strategy",
		"bullets":	["Explain how WSS differs from WS.", "Show the server configuration using a real code snippet.", "Include a list of common certificate pitfalls."]
	}, {
		"title":	"Desired Outcome",
		"body":	"The developer should be able to confidently deploy a WSS server with correct cert configuration."
	}, {
		"title":	"Constraints",
		"bullets":	["Limit to 300 words.", "Assume they use Node.js with the ws library.", "Avoid unrelated technologies like NGINX or Apache."]
	}]
Rendered Markdown:


## System Persona

Act as a security-focused backend engineer experienced with WebSockets and TLS.

## Problem Context

A developer is trying to set up a secure WebSocket (wss://) server but is unsure about certificate handling and client validation.

## Response Strategy

- Explain how WSS differs from WS.
- Show the server configuration using a real code snippet.
- Include a list of common certificate pitfalls.

## Desired Outcome

The developer should be able to confidently deploy a WSS server with correct cert configuration.

## Constraints

- Limit to 300 words.
- Assume they use Node.js with the ws library.
- Avoid unrelated technologies like NGINX or Apache.
Rendered XML:

<prompt>
  <section>
    <title>System Persona</title>
    <body>Act as a security-focused backend engineer experienced with WebSockets and TLS.</body>
  </section>
  <section>
    <title>Problem Context</title>
    <body>A developer is trying to set up a secure WebSocket (wss://) server but is unsure about certificate handling and client validation.</body>
  </section>
  <section>
    <title>Response Strategy</title>
    <bullets>
      <bullet>Explain how WSS differs from WS.</bullet>
      <bullet>Show the server configuration using a real code snippet.</bullet>
      <bullet>Include a list of common certificate pitfalls.</bullet>
    </bullets>
  </section>
  <section>
    <title>Desired Outcome</title>
    <body>The developer should be able to confidently deploy a WSS server with correct cert configuration.</body>
  </section>
  <section>
    <title>Constraints</title>
    <bullets>
      <bullet>Limit to 300 words.</bullet>
      <bullet>Assume they use Node.js with the ws library.</bullet>
      <bullet>Avoid unrelated technologies like NGINX or Apache.</bullet>
    </bullets>
  </section>
</prompt>