smite: add is_standard_shutdown_script helper by ekzyis · Pull Request #186 · lnfuzz/smite

ekzyis · 2026-08-03T16:36:19Z

Depends on #192.

From the commit message:

BOLT-02 specifies sender requirements for shutdown scripts. They must be
witness v0 (P2WPKH, P2WSH) or following features must be negotiated:

option_shutdown_anysegwit: witness v1-v16 with a 2..=40 byte program

option_simple_close: OP_RETURN with a single data push of 6..=80
bytes

Legacy scripts (P2PKH, P2SH) may be accepted for backward compatibility.

This applies to the shutdown and closing_complete messages, and the
upfront_shutdown_script TLV in the open_channel, open_channel2,
accept_channel and accept_channel2 messages.

This commit adds a helper to catch targets that don't comply with the
spec.

~~As per the note I added to the code, I'm not sure if the fuzzer should also reject legacy scripts, since a target must not send them.~~ update: decided to reject them, see discussion

I haven't wired this into existing code or #163 yet, but I thought the introduction of the helper might be worthwile to review itself, especially considering the question wrt legacy scripts.

NishantBansal2003

Thanks! I was about to add this as a follow-up to #185, but it looks like I don’t have to now

NishantBansal2003

Mostly, it looks good. Just a few small comments. I'm not sure how this follows the merge with #192, but if it's merged before, I'll add the updated feature check there

NishantBansal2003 · 2026-08-11T13:32:31Z

+/// Returns `true` if `spk` is a standard `shutdown` scriptpubkey per BOLT 2: P2WPKH or P2WSH.
+/// Negotiated `features` widen the accepted set.
+///
+/// Legacy P2PKH/P2SH are rejected. A receiver may accept them for backward compatibility, but this


In the oracle, we verify what the peer accepted (and whether it was valid), and then we check our side to ensure that the target is actually sending that.

So, when verifying a target-accepted open_channel, I need it to pass even if it contains the legacy script. However, for our received accept_channel, I need to ensure that it does not contain legacy scripts

I didn't address this yet in 63fed11. I think we want to handle this in a similar way to enum Side.

NishantBansal2003 · 2026-08-11T13:42:46Z

+
+    #[test]
+    #[allow(clippy::similar_names)]
+    fn is_standard_shutdown_script_rejects_legacy_accepts_witness_v0() {


nit: I think for testing semantic script types, we should use the bitcoin crate to create the scripts, eg: ScriptBuf::new_p2wpkh(&WPubkeyHash::all_zeros()).into_bytes();, but for testing non-standard cases, we can construct the raw bytes manually

Mhh. If we used bitcoin::ScriptBuf to construct standard scripts via ScriptBuf::new_p2wpkh etc., I think we'd mostly be testing the bitcoin crate against itself, since is_standard_shutdown_script is implemented with bitcoin::Script::is_p2wpkh().

It would also make it less clear how we are testing. I think raw opcodes make it explicit which bytes we are testing. I think tests should be optimized for clear understanding of what we are testing how. This is why I think it's ok to use bitcoin::opcodes.

I think @morehouse mentioned something similar in a review, but I didn't find it.

ekzyis · 2026-08-13T11:52:56Z

I'm not sure how this follows the merge with #192

Putting this into draft until #192 is merged, I rebased this and I addressed the feedback

ekzyis · 2026-08-20T15:53:07Z

I rebased this PR on top of #192 since I don't expect it to change much. I didn't address all feedback yet, see #186 (comment) and TODO in the code + commit message.

Update: rebased again on top of #192 (ACK 40f7aa8)

ekzyis · 2026-08-24T14:41:29Z

TODO:

support legacy shutdown scripts for open_channel sent to target (smite: add is_standard_shutdown_script helper #186 (comment))

morehouse

Great work! Needs a rebase now that #192 is merged. Also would be good to include the additions from #209 before merging this.

ekzyis · 2026-09-01T21:22:42Z

c62c850: rebased on master (87ebab6), added is_acceptable_shutdown_script from #192, and updated commit message. I think is_acceptable_shutdown_script is simple and enough for our needs; no need for enum Side like we did for commitments.

f024116: fixed wrong usage of backticks in is_standard_shutdown_script documentation

BOLT-02 specifies sender requirements for shutdown scripts. They must be witness v0 (P2WPKH, P2WSH) or following features must be negotiated: * `option_shutdown_anysegwit`: witness v1-v16 with a 2..=40 byte program * `option_simple_close`: `OP_RETURN` with a single minimal data push of 6..=80 bytes This applies to the `shutdown` and `closing_complete` messages, and the `upfront_shutdown_script` TLV in the `open_channel`, `open_channel2`, `accept_channel` and `accept_channel2` messages. This commit adds two helpers to catch targets that don't comply with the spec: is_standard_shutdown_script and is_acceptable_shutdown_script.

morehouse · 2026-09-01T22:31:26Z

+pub fn is_acceptable_shutdown_script(spk: &[u8], features: &Features) -> bool {
+    let script = Script::from_bytes(spk);
+    is_standard_shutdown_script(spk, features) || script.is_p2pkh() || script.is_p2sh()
+}


We should add tests for is_acceptable_shutdown_script.

morehouse · 2026-09-01T22:40:21Z

+        //
+        // TODO: Oracle verification depends on if we're verifying a target's message or our own
+        // message. Legacy scripts MAY be accepted by receivers, but MUST NOT be sent.


Suggested change

//

// TODO: Oracle verification depends on if we're verifying a target's message or our own

// message. Legacy scripts MAY be accepted by receivers, but MUST NOT be sent.

ekzyis force-pushed the is-standard-shutdown-script branch from f33051b to 22217d9 Compare August 4, 2026 11:07

ekzyis mentioned this pull request Aug 4, 2026

smite: add accept_channel oracle #185

Merged

NishantBansal2003 reviewed Aug 9, 2026

View reviewed changes

Comment thread smite/src/bolt/shutdown.rs Outdated

Comment thread smite/src/bolt/shutdown.rs Outdated

Comment thread smite/src/bolt/shutdown.rs Outdated

Comment thread smite/src/bolt/shutdown.rs Outdated

ekzyis force-pushed the is-standard-shutdown-script branch from 22217d9 to 220b1bf Compare August 9, 2026 12:30

ekzyis commented Aug 9, 2026

View reviewed changes

Comment thread smite/src/bolt/shutdown.rs Outdated

ekzyis force-pushed the is-standard-shutdown-script branch from 220b1bf to 847c2f1 Compare August 9, 2026 18:00

NishantBansal2003 approved these changes Aug 11, 2026

View reviewed changes

ekzyis marked this pull request as draft August 13, 2026 11:53

NishantBansal2003 mentioned this pull request Aug 19, 2026

smite: validate negotiated features and channel types in accept_channel oracle #209

Draft

11 tasks

ekzyis mentioned this pull request Aug 20, 2026

smite-ir: add RecvShutdown operation #163

Draft

3 tasks

ekzyis force-pushed the is-standard-shutdown-script branch 2 times, most recently from dd42b6b to 63fed11 Compare August 20, 2026 15:47

ekzyis force-pushed the is-standard-shutdown-script branch from 63fed11 to 6b8d367 Compare August 21, 2026 15:16

ekzyis marked this pull request as ready for review August 24, 2026 14:06

ekzyis mentioned this pull request Aug 24, 2026

refactor: test: add recv_channel_ready_instructions helper #213

Open

ekzyis marked this pull request as draft August 24, 2026 14:28

ekzyis marked this pull request as ready for review August 24, 2026 14:31

ekzyis marked this pull request as draft August 24, 2026 14:37

morehouse reviewed Sep 1, 2026

View reviewed changes

ekzyis force-pushed the is-standard-shutdown-script branch from 6b8d367 to c62c850 Compare September 1, 2026 21:18

ekzyis marked this pull request as ready for review September 1, 2026 21:23

ekzyis force-pushed the is-standard-shutdown-script branch from c62c850 to f024116 Compare September 1, 2026 21:28

morehouse reviewed Sep 1, 2026

View reviewed changes

	//
	// TODO: Oracle verification depends on if we're verifying a target's message or our own
	// message. Legacy scripts MAY be accepted by receivers, but MUST NOT be sent.

Uh oh!

Conversation

ekzyis commented Aug 3, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

NishantBansal2003 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

NishantBansal2003 left a comment

Choose a reason for hiding this comment

Uh oh!

NishantBansal2003 Aug 11, 2026

Choose a reason for hiding this comment

Uh oh!

ekzyis Aug 20, 2026

Choose a reason for hiding this comment

Uh oh!

NishantBansal2003 Aug 11, 2026

Choose a reason for hiding this comment

Uh oh!

ekzyis Aug 20, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

ekzyis commented Aug 13, 2026

Uh oh!

ekzyis commented Aug 20, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

ekzyis commented Aug 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

morehouse left a comment

Choose a reason for hiding this comment

Uh oh!

ekzyis commented Sep 1, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

morehouse Sep 1, 2026

Choose a reason for hiding this comment

Uh oh!

morehouse Sep 1, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

ekzyis commented Aug 3, 2026 •

edited

Loading

ekzyis commented Aug 20, 2026 •

edited

Loading

ekzyis commented Aug 24, 2026 •

edited

Loading

ekzyis commented Sep 1, 2026 •

edited

Loading