configgle deserializes configuration into live Python objects, and doing so imports the modules a config names. A config file is therefore closer to code than to data: deserializing one that an attacker controls can import arbitrary importable modules and reach their import-time side effects. Security reports need a private path so exploit details are not published before review.

Please report suspected security vulnerabilities privately by emailing hello@rekursiv.ai.

Include:

• Affected version or commit.
• Steps to reproduce.
• Expected impact.
• Any suggested mitigation.

Please do not open public issues for vulnerabilities until we have investigated and coordinated disclosure.

Security reports are especially useful for:

• Deserialization that resolves a dotted path to an object and imports it (Fig.deserialize and configgle.lib.codec.from_plain), including any way to widen what a config can name.
• Config sources that reach the deserializer without being trusted first -- files, environment, or command-line overrides.
• Dependency or packaging issues that affect installed users.
• Supply-chain concerns in the published wheel or its dependency set.

Deserializing a config you do not trust is equivalent to running code you do not trust; treat config files with the same care as source files. Reports that amount to "an untrusted config can import a module" describe the documented behavior above rather than a vulnerability.