Skip to content
IAM Insights

Explainers · Glossary

DNSSEC

DNS Security Extensions

DNS Security Extensions: cryptographic signatures on DNS answers so a resolver can tell a genuine record from a forged one.

In our briefings

The claim each of the newest briefings on DNSSEC would have you leave with.

  1. Revision -02 of Shankarlingam's draft says the delay before a disabled agent is refused is "a property of the receiver's configuration, not of the token alone", and that passing the receiver's checks does not show a disabled client is refused within the bound the receiver computes. Raising the maximum token age, DNS caching, caching of issuer metadata and keys, or the clock-skew allowance lengthens the time a disabled client is still accepted, while every signature, issuer, client, audience and DNS check keeps passing. Those checks, the draft says, "establish that a token is authentic; they do not establish that the client is still enabled." A receiver SHOULD recompute the bound before any change to those settings takes effect, and SHOULD test it directly, again after any change to those settings: obtain link tokens for a test client, disable it at a time not known to whoever observes the receiver, keep presenting tokens issued before the disable, and compare the last acceptance with the time the authorization server records the disable. The receiver meets its bound if that gap does not exceed the computed delay. It SHOULD also log the maximum age in effect with each verification result, and the security section says receivers SHOULD treat such changes as security relevant. The draft describes this test and reports no results from running it.

    A disabled agent's old tokens still pass every check; a DNS draft now asks receivers to time how longBI-FIND-2026-2345

  2. Revision -01, published on 27 October, puts a time limit on the Authority's suspensions and emergency steps. Staff may suspend a registrar's new enrollments and a board majority its issuer rights, but a suspension now lapses after ninety days unless two-thirds of the board extend it, 'so a suspension cannot become a revocation by delay.' The security function, which -01 defines as a standing group of at least three people, none employed by or affiliated with an accredited party or an included manufacturer, may suspend a registrar at once on credible evidence that it is compromised. That suspension lasts at most thirty days and lapses unless the board ratifies it by a two-thirds vote. An emergency distrust of a manufacturer's keys likewise lapses at the end of the thirtieth day if the board has not acted, and the declaration must state the evidence for any cutoff date and its uncertainty: it 'MUST NOT claim a precision it lacks.' Revocation still takes two-thirds of the board. The -00 text set no time limit on a suspension and gave no end for an emergency step the board left unratified.

    Drake's revised agent registry charter makes suspensions lapse at 90 days and still bars erasing an identityBI-FIND-2026-1833

  3. Drake's resolution draft has a verifier trust an agent's token only after the registry says who may issue it. The agent's identifier becomes a DNS lookup under urn.arpa, which leads to an RDAP record carrying currentIssuer, the issuer URI of the agent's current registrar. If the record has none, the verifier must reject registrar-backed authentication; if it has one, the token's iss must equal it, since a token's own iss "MUST NOT bootstrap its own authority." DNSSEC is mandatory in production, but the urn.arpa entry is not yet registered, so for now a deployment may point at the interim 1id.biz resolver or registry host as a local trust decision. The example RDAP response in revision -01 is a live answer from that reference service on 2026-10-22, and its currentIssuer is a URI on 1id.com, the author's company. Our briefing on Drake's governance draft found that he and 1id.com run the reference registry, so we read the example as the author demonstrating his own service; the -01 text names no other implementation.

    Drake's agent registry names who may issue an agent's token, and its answer has no signature of its ownBI-FIND-2026-0751

›Cited in 8 briefings

Every briefing that discusses this term, newest first.

  1. A disabled agent's old tokens still pass every check; a DNS draft now asks receivers to time how long

    Draft · · IETF Datatracker · Sriram Shankarlingam, HexaEight · Oct 20, 2026

  2. Drake's revised agent registry charter makes suspensions lapse at 90 days and still bars erasing an identity

    Draft · · IETF Datatracker · Christopher Drake, 1id.com · Sep 25, 2026

  3. Drake's agent registry names who may issue an agent's token, and its answer has no signature of its own

    Draft · · IETF Datatracker · Christopher Drake, 1id.com · Sep 25, 2026

  4. OAuth tokens prove an agent's rights, not a person's consent; revision 05 adds a day-long grant with spend caps

    Draft · · IETF Datatracker · Meiling Chen et al. · Oct 27, 2026

  5. The Agent Name Service draft was rebuilt on DNS, with a GoDaddy author, and no standards body has adopted it

    Draft · · IETF Datatracker · Scott Courtney et al. · Apr 13, 2026

  6. An A2A threat model flags the Agent Card as an attack path, and its example flow never passes on the user's identity

    Paper · · arXiv · Idan Habler et al. · Apr 23, 2025

  7. FAPI 2.0 issues only sender-constrained access tokens and leaves public clients out

    Article · · OpenID Foundation · Daniel Fett et al. · Feb 22, 2025

  8. DANE lets a domain vouch for its own TLS keys in DNS, and a bad record locks out every client that checks

    Article · · RFC Editor · Paul Hoffman et al. · Aug 7, 2012