psbt: Fix PSBTInputSignedAndVerified bounds assert #34272

1. 
   l0rinc commented at 12:01 pm on January 13, 2026: contributor

   This PR fixes an off-by-one in a debug assertion in PSBTInputSignedAndVerified. The function indexes psbt.inputs[input_index], so the assertion must not allow indexing at psbt.inputs.size().

   Found during review: #31650 (review)

2. psbt: Fix `PSBTInputSignedAndVerified` bounds `assert`

   The previous `assert` used `>=`, allowing `input_index == psbt.inputs.size()` and out-of-bounds access in `psbt.inputs[input_index]`.
   
   Found during review: https://github.com/bitcoin/bitcoin/pull/31650#discussion_r2685892867

   2f5b1c5f80
3. DrahtBot added the label PSBT on Jan 13, 2026
4. 
   DrahtBot commented at 12:01 pm on January 13, 2026: contributor

   The following sections might be updated with supplementary metadata relevant to reviewers and maintainers.

   Code Coverage & Benchmarks

   For details see: https://corecheck.dev/bitcoin/bitcoin/pulls/34272.

   Reviews

   See the guideline for information on the review process.

   Type	Reviewers
   ACK	optout21, maflcko, achow101

   If your review is incorrectly listed, please copy-paste <!–meta-tag:bot-skip–> into the comment that the bot should ignore.

5. 
   optout21 commented at 1:40 pm on January 13, 2026: contributor

   utACK 2f5b1c5f80590ffa6b5a5bcfb21fddb1dc22e852 Trivial change to an assert, can be reliably assessed by local code inspection (2 lines). Code reviewed; build & unit tests verified locally.
6. 
   maflcko commented at 2:08 pm on January 13, 2026: member

   lgtm ACK 2f5b1c5f80590ffa6b5a5bcfb21fddb1dc22e852

   This is just a refactor/doc change, because the UB can not be reached in the current code-base, and is assumed to be unreachable anyway (due to the use of assert)

7. maflcko added the label Refactoring on Jan 13, 2026
8. 
   willcl-ark commented at 2:58 pm on January 13, 2026: member

   FWIW I think I hit this on the fuzz tests here: https://github.com/willcl-ark/bitcoin/actions/runs/20883835381/job/60004100089#step:10:5614 So it may be possible for this to cause CI failure, at least.
9. 
   maflcko commented at 3:09 pm on January 13, 2026: member

   @willcl-ark No, that is #33999 from the input fuzz_corpora/psbt/3fa30f92df4e391124a56b76cc3db3eb71b5d69c from commit https://github.com/bitcoin-core/qa-assets/pull/252/changes/00c335ca2ac2831fdf3d0fc2197ac509530ae13b
10. 
    willcl-ark commented at 3:33 pm on January 13, 2026: member

    @willcl-ark No, that is #33999 from the input fuzz_corpora/psbt/3fa30f92df4e391124a56b76cc3db3eb71b5d69c from commit bitcoin-core/qa-assets@00c335c

    ah ok i see, thanks.

11. 
    achow101 commented at 0:15 am on January 14, 2026: member

    ACK 2f5b1c5f80590ffa6b5a5bcfb21fddb1dc22e852
12. achow101 merged this on Jan 14, 2026
13. achow101 closed this on Jan 14, 2026

    ---

14. fanquake referenced this in commit 7e1090f5e8 on Jan 14, 2026
15. 
    fanquake commented at 5:27 pm on January 14, 2026: member

    Backported to 30.x in #34283.

Contributors
l0rinc DrahtBot optout21 maflcko willcl-ark achow101 fanquake

Labels
Refactoring PSBT