[media-types] Re: [IANA #1456754] application/vnd.no mos registration request
"Murray S. Kucherawy" <[email protected]>
| Newsgroups | gmane.ietf.types |
|---|---|
| Message-ID | <CAL0qLwZLLfQLhWvGrmx19u86=koN4ttxT_waob1m5Tj3Fht6DA@mail.gmail.com> |
A couple of minor things here. -MSK On Mon, Jul 27, 2026 at 6:00 PM David Dong via RT < [email protected]> wrote: > Hi Murray, > > Can you review this new request by August 10th? > > Thank you. > > -- > > Name: Allan Sendagi > > Email: [email protected] > > Media type name: application > > Media subtype name: vnd.nomos > > Required parameters: N/A > > > > Optional parameters: N/A > > > > Encoding considerations: binary > > The .nomos file is a cryptographic container. It can contain arbitrary > binary data, and as such, selecting binary ensures it can be safely > transmitted without corruption over any transport. > > Security considerations: The .nomos file is a governance artifact > containing cryptographically sealed rules and policies. It is not > executable code in the traditional sense; rather, it is a data structure > evaluated by a runtime kernel (ATLAS) to govern agent actions. The security > of the artifact is rooted in its cryptographic seal, which ensures > integrity and authenticity. It does not contain active content. The > evaluation engine is designed to be deterministic and operates under a > "deny-by-default" policy, ensuring that any action not explicitly permitted > is blocked. All governance decisions are fully auditable. The format does > not inherently provide privacy services, so it should be transmitted over > secure channels (e.g., TLS/HTTPS) when necessary. > I think the second sentence refers specifically to integrity, not security generally, because the last sentence states that privacy is not assured. Might want to be specific. > Interoperability considerations: The .nomos artifact is a portable, sealed > governance file consumed by the nomos runtime. Interoperability is achieved > through a published specification. The evaluation of its contents is > deterministic, ensuring that the same .nomos file yields the same results > on any compatible runtime. > Are there any prior versions of this format out there that might be incompatible with this specification? > Published specification: https://github.com/project-nomos/spec (This is > the link to the .nomos specification repository). > > > > Applications which use this media: The .nomos format is used by the NOMOS > protocol to package, distribute, and enforce machine-readable authority for > AI agents. It is consumed by the runtime kernel to mediate actions and > ensure compliance with defined policies. > > Fragment identifier considerations: N/A > > > > Restrictions on usage: N/A > > > > Provisional registration? (standards tree only): Yes > Should be "N/A" as this is not a standards tree registration. > Additional information: > > 1. Deprecated alias names for this type: N/A > 2. Magic number(s): 0x4E 0x4F 0x4D > 3. File extension(s): .nomos > 4. Macintosh file type code: N/A > 5. Object Identifiers: N/A > > General Comments: > > Person to contact for further information: > > 1. Name: Allan Sendagi > 2. Email: [email protected] > > Intended usage: COMMON > > Author/Change controller: SAFEHAVEN LLC > _______________________________________________ media-types mailing list -- [email protected] To unsubscribe send an email to [email protected]