dillon-b

[mintlify]

Mintlify

0 threads from 0 users in Mintlify

• No unresolved comments

[greptile-apps]

Confidence Score: 4/5

The PR should not merge until the changelog accurately documents boolean Runtime Property targeting.

The previously reported omission remains: the changelog presents an exhaustive data-type list without the documented True/False type or its dedicated operators.

Files Needing Attention: changelogs.mdx

Important Files Changed

Filename	Overview
changelogs.mdx	Adds two product announcements covering plan limits, experimentation capabilities, and typed feature-flag Runtime Properties.

_{Reviews (2): Last reviewed commit: "Updated mintlify pages" | Re-trigger Greptile}

[greptile-apps]

Boolean data type omitted

The exhaustive list omits the dedicated True/False data type and its is True and is False operators, leading readers to infer that boolean properties require fuzzy matching.

Suggested change

	Feature Flags can now target Runtime Properties by their actual data type — string, number, list, or fuzzy — instead of treating every value as a string. Targeting rules match correctly on numbers and lists, not just text.
	Previously, Runtime Property filters compared everything as a string. That meant no real numeric or boolean comparisons, making flag targeting less predictable and reliable.
	With this update, each targeting condition has a Data Type that controls how it's parsed and compared:
	- **String:** case-insensitive match, ordered comparisons in lexicographical order
	- **Number:** true numeric comparison (Is, Greater than, Less than, etc.)
	- **List:** match any/all elements, each evaluated by its own type
	- **Fuzzy:** loose, cross-type match (e.g. `"5"` matches `5`, `"1"` matches `true`)
	Feature Flags can now target Runtime Properties by their actual data type — string, number, true/false, list, or fuzzy — instead of treating every value as a string. Targeting rules match correctly on numbers, booleans, and lists, not just text.
	Previously, Runtime Property filters compared everything as a string. That meant no real numeric or boolean comparisons, making flag targeting less predictable and reliable.
	With this update, each targeting condition has a Data Type that controls how it's parsed and compared:
	- **String:** case-insensitive match, ordered comparisons in lexicographical order
	- **Number:** true numeric comparison (Is, Greater than, Less than, etc.)
	- **True/False:** boolean comparison using Is True or Is False
	- **List:** match any/all elements, each evaluated by its own type
	- **Fuzzy:** loose, cross-type match (e.g. `"5"` matches `5`, `"1"` matches `true`)