Skip to content

chore(vcr-mem): re-validate scry consumed surface against scry-sai-core 2.3.0 (#383, #242)#523

Merged
avrabe merged 1 commit into
mainfrom
vcr-mem-scry-2.3.0-revalidate-242
Jun 27, 2026
Merged

chore(vcr-mem): re-validate scry consumed surface against scry-sai-core 2.3.0 (#383, #242)#523
avrabe merged 1 commit into
mainfrom
vcr-mem-scry-2.3.0-revalidate-242

Conversation

@avrabe

@avrabe avrabe commented Jun 27, 2026

Copy link
Copy Markdown
Contributor

What

scry released v2.3.0. synth's VCR-MEM-001 (#383) layer-2 substrate pins the
scry-sai-core dev-dependency at "2.0" with a comment asserting the 2.x bump is
"transparent" (consumed surface unchanged) — but that was only ever verified against
2.0.0 (the lockfile pin), untested for 2.1/2.2/2.3.

This bumps the lockfile to scry-sai-core 2.3.0 and re-runs the in-tree integration
test (scry_shadow_stack_budget) that calls scry_analyze_core::analyze() on the real
gust-family module. GREEN — the consumed surface is unchanged:

  • stack_usage.max_stack_bytes = 65536
  • function_summaries[].recursive (honest-fail recursion gate)
  • reachable_from_exports (dead-path pruning superset)

So the "transparent bump" claim is now empirically backed, not just asserted. Doc
comments updated to record the 2.3.0 re-verification.

Gate

Behavior-frozen — scry-sai-core is a dev-dependency exercised only by this test, so
the production synth binary does not pull scry: no codegen / .text change. Frozen
anchors bit-identical (frozen_codegen_bytes 3/3, control_step 0x00210A55 / flight_algo
0x07FDF307 untouched); cargo fmt --check clean.

Responds to the scry v2.3.0 release; the directive's preferred "external-analysis
dev-dependency + test" shape.

🤖 Generated with Claude Code

…re 2.3.0 (#383, #242)

scry released v2.3.0; synth's VCR-MEM-001 layer-2 substrate pins the scry-sai-core
DEV-dependency at `"2.0"` (caret) with a comment asserting the 2.x bump is
"transparent" because the consumed surface is unchanged. That claim was only
verified against 2.0.0 (the lockfile pin) — untested for 2.1/2.2/2.3.

Bump the lockfile to scry-sai-core 2.3.0 and re-run the in-tree integration test
(scry_shadow_stack_budget) that analyzes the real gust-family module: GREEN — the
consumed surface (stack_usage.max_stack_bytes=65536, function_summaries[].recursive,
reachable_from_exports) is unchanged, so the "transparent" claim is now empirically
backed, not just asserted. Doc comments updated to record the 2.3.0 re-verification.

Behavior-frozen: scry-sai-core is a DEV-dependency exercised only by this test — the
production `synth` binary does not pull scry, so no codegen / `.text` change. Frozen
anchors bit-identical (frozen_codegen_bytes 3/3); fmt clean.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
@codecov

codecov Bot commented Jun 27, 2026

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

@avrabe avrabe merged commit 8f2fca4 into main Jun 27, 2026
24 checks passed
@avrabe avrabe deleted the vcr-mem-scry-2.3.0-revalidate-242 branch June 27, 2026 01:41
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