Merge branch 'github:main' into fix/path-injection-read-subkind

This commit is contained in:
Kaixuan Li
2026-04-21 22:59:53 +10:00
committed by GitHub
197 changed files with 1499 additions and 623 deletions

View File

@@ -40,14 +40,26 @@ string getCommonSensitiveInfoRegex() {
/**
* Gets a regular expression for matching common names of variables that
* indicate the value being held does not contains sensitive information,
* indicate the value being held does not contain sensitive information,
* but is a false positive for `getCommonSensitiveInfoRegex`.
*
* - "tokenizer" is often used for java.util.StringTokenizer.
* - "tokenImage" appears in parser code generated by JavaCC.
* - Pagination/iteration tokens: "nextToken" (AWS SDK), "pageToken" (GCP), etc.
* - Token metadata: "tokenType" (OAuth), "tokenEndpoint" (OIDC), "tokenCount", etc.
* - Secret metadata: "secretName" (K8s/AWS), "secretId" (Azure), "secretVersion", etc.
*/
string getCommonSensitiveInfoFPRegex() {
result = "(?i).*(null|tokenizer).*" or result = "tokenImage"
result =
[
"(?i).*(null|tokenizer).*", "tokenImage",
// Pagination/iteration tokens (e.g., AWS SDK pagination cursors, parser tokens)
"(?i).*(next|previous|current|page|continuation|cursor)tokens?.*",
// Token metadata/infrastructure (token followed by a non-value descriptor)
"(?i).*tokens?(type|kind|count|index|position|length|offset|endpoint|url|uri|bucket|rate|delimiter|separator|format|number|name|id|prefix|suffix|pattern|class|style).*",
// Secret metadata (secret followed by a non-value descriptor)
"(?i).*secrets?(name|id|version|ref|arn|path|type|label|description|manager|client|provider|store|factory|properties).*"
]
}
/** An expression that might contain sensitive data. */