Skip to main content
Agents can report one or more credentials in any post_response message. Mythic 4.0 stores a credential type and subtype, parses supported formats, tracks structured identity and metadata, and automatically evaluates time-based validity.

Agent response

credential_type is normalized to one of:
  • plaintext
  • certificate
  • hash
  • key
  • ticket
  • cookie
  • hex
  • jwt
Unknown types fall back to plaintext. credential_subtype is a normalized, free-form value that distinguishes formats within a type, such as a particular hash algorithm or token role. Mythic requires at least an account, a realm, or a non-empty credential_identity. The remaining enrichment fields are optional:
  • comment records operator-facing context.
  • custom_display overrides the compact label used in the UI.
  • metadata stores descriptive and lifecycle data.
  • credential_identity stores stable structured identity fields.
Any additional keys supplied on the credential object are merged into metadata for compatibility.

Parsing and validity

Mythic includes parsers for structured formats such as JWT and Kerberos credentials. A parser can populate the subtype, identity, account, realm, and metadata from the credential value. Explicit structured values are preserved where appropriate. Lifecycle metadata uses RFC 3339 timestamps:
Mythic calculates a validity object inside the stored metadata and maintains operation tags when the credential is not yet valid, expired, or past its renewal window. When the same credential, type, subtype, identity, account, and realm are reported again in an operation, Mythic refreshes the existing record instead of creating a duplicate. See Credentials for operator workflows and Tasking References for using @cred references in task arguments.