(registered 2026-09-15, last updated 2026-09-15)

Scheme name:
ace

Status:
Provisional

Applications/protocols that use this scheme name:
The "ace" URI scheme identifies governed ACE objects in a
transport-independent namespace. An "ace" URI identifies an object or
protocol resource; it does not by itself establish verification,
authority, permission to act, execution, observation, or settlement.
Applications parse and validate ace URIs, compare them for equivalence,
request discovery or resolution of the identified object, retrieve a
candidate public representation or receipt, verify a representation
using an applicable ACE verification profile, and return machine-readable
refusals when a result cannot be established.

Contact:
Frank Bryan
Seqentra
Sydney, Australia
ip&seqentra.com

Change controller:
Seqentra IP Holdings Pty Ltd
Sydney, Australia
ip&seqentra.com

Syntax:
The scheme conforms to the generic URI syntax in RFC 3986:

ace-URI = "ace:" hier-part [ "?" query ] [ "#" fragment ]

The initial profile uses hierarchical form:

ace://namespace-class/resource-segment *( "/" resource-segment )

Where:

namespace-class = "root" / "ai" / "x" / "kernel" / "cortex" /
"federation" / "agent" / "service" / "repo" /
"capsule"

resource-segment = 1*64(segment-char)
segment-char = ALPHA / DIGIT / "." / "_" / "~" / "-"

The external profile is case-sensitive after the scheme name. Producers
MUST emit lowercase namespace classes and lowercase resource segments.
Consumers MUST reject ambiguous or non-canonical forms rather than
silently rewriting identity-bearing components.

Examples:
ace://agent/delta
ace://kernel/current
ace://cortex/current
ace://federation/current
ace://repo/example-project
ace://capsule/example-object

Encoding considerations / Internationalisation:
The initial profile limits namespace classes and resource segments to
ASCII. Non-ASCII names MUST be represented through a separately defined
encoding or mapping profile before being placed in an ace URI.
Implementations MUST NOT apply locale-dependent case conversion or
Unicode normalization to canonical identity components. Scheme names are
compared case-insensitively under RFC 3986; all remaining identity-bearing
components are compared octet-for-octet after URI syntax validation and
permitted percent-encoding normalization.

Interoperability considerations:
The scheme follows RFC 3986 generic syntax. Canonical identity-bearing
components are compared octet-for-octet after permitted URI normalization.
The scheme is transport independent. External aliases, projections,
redirects or compatibility mappings do not create a second canonical
identity. Implementations must preserve the distinction between
identifier, discovered representation, verified representation, authority,
action, observation and settlement.

Security considerations:
Implementations must treat resolution as discovery only, and must fail
closed where a result is malformed, ambiguous, contradictory, stale,
revoked, expired or unverifiable. A registry, resolver, gateway, redirect,
transport or hosting provider does not create authority. Implementations
must not represent discovery as successful verification. Implementations
should minimise disclosure of private identifiers, internal topology,
custody information and non-public metadata. An ace URI must not embed
secrets, private keys, bearer credentials or sensitive personal
information.

References:
- RFC 3986, Uniform Resource Identifier (URI): Generic Syntax.
- RFC 7595, Guidelines and Registration Procedures for URI Schemes.
- Provisional public specification: https://seqentra.com/specifications/ace-uri-scheme