Skip to content

docs(cbom): explain how to read and report nistQuantumSecurityLevel - #131

Open
Lennonhaha wants to merge 1 commit into
CycloneDX:mainfrom
Lennonhaha:docs/nistqsl-guidance
Open

Lennonhaha wants to merge 1 commit into
CycloneDX:mainfrom
Lennonhaha:docs/nistqsl-guidance

Conversation

@Lennonhaha

Copy link
Copy Markdown

Adds the consumer/producer guidance for nistQuantumSecurityLevel to the CBOM guide, so it does not have to live in the specification schema description (CycloneDX/specification#1177).

What the section covers

  • absence and 0 are different statements, and absence must not be read as 0
  • absence is not evidence that an asset is quantum-safe
  • in Protocol Buffers the field is optional int32, so generated accessors return 0 for an unset field — check presence (has_nist_quantum_security_level()) before comparing the value
  • when to emit 0 (assessment performed, no category met) and when to omit the property (no assessment, or unknown result)

Placement

CBOM/en/0x30-Anatomy-of-a-CBOM.md, new ## Security Level Properties section between the asset-type table and ## Key Management. It covers classicalSecurityLevel as well, since both are properties of the algorithm rather than of a deployment.

Not included

No statement about which category number maps to which algorithm family — that belongs to the algorithm definitions and the NIST documents themselves.

Companion change: CycloneDX/specification#1177 (schema description narrowed to the definition and the meaning of an explicit zero).

The interpretation guidance for this property was removed from the
specification schema description at the code owner's request
(CycloneDX/specification#1177): the spec keeps the definition and the
meaning of an explicit zero, while the guide carries the guidance about
how the value is consumed and produced.

Covers:
- absence and `0` are different statements, and absence is not `0`
- in Protocol Buffers the field is `optional int32`, so the generated
  accessor returns 0 for an unset field: check presence first
- when to emit `0` and when to omit the property

Signed-off-by: 刘天赫 <research@fibemate.net>
@Lennonhaha
Lennonhaha force-pushed the docs/nistqsl-guidance branch from d2fab0f to f0948d1 Compare November 15, 2026 06:31
@Lennonhaha

Copy link
Copy Markdown
Author

DCO is green on f0948d1 now. The failure was not a missing -s: the commit carried Signed-off-by: 刘天赫 <research@fibemate.net> from the start, but the commit author was liuguowei1983 <27202998@qq.com> — this clone had no local git identity, so it fell back to the global one. DCO compares the sign-off against the commit author, so a sign-off line alone is not enough. Fixed with git commit --amend --reset-author --no-edit, which keeps the message and makes the identity match.

Two things still stand between this and a merge, and neither is something I can do from here:

  • the Zizmor workflow on this head is action_required — fork PRs need a maintainer to approve the run, so its check stays pending until someone clicks approve.
  • the PR needs a review; there are none yet.

Content is unchanged: one file, +55/−0, the Security Level Properties section inserted before ## Key Management in CBOM/en/0x30-Anatomy-of-a-CBOM.md, CRLF preserved. Happy to adjust wording or placement if you would rather see it elsewhere in the guide.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant