Issues

• cargo-deny reddens main on a roughly weekly cadence, for a wasmtime exposure scry does not have

  pulseengine/scry#230 · Filter by author avrabe opened on Oct 21, 2026
  1

• Linux CI jobs still run on hosted runners while the org fleet sits idle — the part of #205 that survives

  pulseengine/scry#227 · Filter by author avrabe opened on Oct 14, 2026
  1

• Two-local unsigned guards (index <u len) are the dominant bounds-check idiom — and the one nothing refines

  pulseengine/scry#226 · Filter by author avrabe opened on Sep 23, 2026

• Doc and artifact accuracy: an unenforced soundness invariant, an unfalsifiable accepted AC, and three consumer-facing overstatements

  pulseengine/scry#220 · Filter by author avrabe opened on Sep 22, 2026

• Four smaller gate holes: a swallowed WIT check, an evadable count-max, a per-commit opt-out, and an ordering check blind to unscheduled work

  pulseengine/scry#219 · Filter by author avrabe opened on Sep 22, 2026

• Dead code and dead tests: scry-mcdc's only test never runs, pentagon::upper_bound has no caller

  pulseengine/scry#218 · Filter by author avrabe opened on Sep 22, 2026

• check-required-checks.py has two blind spots: it reads 2 of 6 workflow files, and its ruleset-reset detector never runs

  pulseengine/scry#217 · Filter by author avrabe opened on Sep 22, 2026

• A malformed provenance section is dropped SILENTLY on the default path, which the docs call the normal entry point

  pulseengine/scry#216 · Filter by author avrabe opened on Sep 22, 2026

• The MC/DC coverage evidence measures a different binary than the one we ship — different compiler, profile, and target

  pulseengine/scry#215 · Filter by author avrabe opened on Sep 16, 2026

• Front-page and dossier claims that nothing gates: sigil evidence, loom consumption, proof count, and a gate that lies about itself

  pulseengine/scry#214 · Filter by author avrabe opened on Sep 16, 2026

• WIT promises caller-controlled widening-threshold; the fixpoint hardcodes 3 and every test passes 3

  pulseengine/scry#213 · Filter by author avrabe opened on Sep 16, 2026

• wasm-lattice: a second, divergent interval algebra that no CI job compiles

  pulseengine/scry#212 · Filter by author avrabe opened on Sep 16, 2026

• All 11 live soundness-oracle tests pass silently if the composed component is missing (fail-open skip)

  pulseengine/scry#211 · Filter by author avrabe opened on Sep 16, 2026

• Soundness-critical transfers with no test: i32_sub, i32_mul, shr_s, sub_wrap_free, mul_wrap_free

  pulseengine/scry#210 · Filter by author avrabe opened on Sep 16, 2026

• claim-check.py fails OPEN when the claims key is absent — one typo disarms a required gate

  pulseengine/scry#209 · Filter by author avrabe opened on Sep 16, 2026

• check-gate-coverage.py counts a COMMENTED-OUT cargo test as coverage — the exact class its rewrite escaped

  pulseengine/scry#208 · Filter by author avrabe opened on Sep 16, 2026

• The Rocq and Verus proof jobs cannot fail anything — the v1.0 'flips to required' never happened

  pulseengine/scry#207 · Filter by author avrabe opened on Sep 16, 2026
  1

• Feedback wanted: needs and wishes for scry

  pulseengine/scry#203 · Filter by author avrabe opened on Sep 2, 2026

• A correct div-by-zero fix is invisible: br_if on i32.eqz gives a disequality, which an interval cannot hold

  pulseengine/scry#165 · Filter by author avrabe opened on Aug 27, 2026
  1

• Loop-bound inference is the single critical path for static WCET on a dissolved OS: 13 of 28 declines, 11 cascades behind it, 3 of 31 functions bounded

  pulseengine/scry#144 · Filter by author avrabe opened on Aug 26, 2026
  7

• No analysis budget: 22% of real-world modules never finish, and an analyzer that doesn't terminate returns nothing at all

  pulseengine/scry#135 · Filter by author avrabe opened on Aug 25, 2026

• func.wast: Internal("i32 binop with single operand") — operand-stack underflow, unmasked by the #125 fix

  pulseengine/scry#128 · Filter by author avrabe opened on Aug 21, 2026

• 97% of real-world modules hit unsoundness-fallback, only 18% of memory-touching modules fully provable — operator ranking, and 579 fallbacks that name no operator

  pulseengine/scry#126 · Filter by author avrabe opened on Aug 21, 2026
  4

• FEAT-065 rework: matching needs content corroboration — 6 wrong-verdict paths found pre-ship

  pulseengine/scry#122 · Filter by author avrabe opened on Aug 11, 2026
  1

• Publish a kilnd-runnable core-module artifact alongside the component

  pulseengine/scry#118 · Filter by author avrabe opened on Aug 8, 2026