Issues

• TRANSPORT-P01 quotes a capture that is not in the repo — and its stated header size contradicts its own field list

  pulseengine/relay#503 · Filter by author avrabe opened on Oct 1, 2026

• v1.141 clean-room review: all five artifacts refuted — the measurements reproduce, the evidence statements do not

  pulseengine/relay#499 · Filter by author avrabe opened on Oct 1, 2026

• v1.140 clean-room review: five blocking findings — ENDURANCE's 1 h gz hold never ran, FLEET-002 cannot fail

  pulseengine/relay#497 · Filter by author avrabe opened on Oct 1, 2026

• The flight seam is the ONE WIT we hand-write — spar cannot express it, and nothing gates its drift

  pulseengine/relay#492 · Filter by author avrabe opened on Sep 30, 2026

• MockPhysics computes specific force BEFORE the ground clamp — the same diff fixed exactly this in SimBackend

  pulseengine/relay#485 · Filter by author avrabe opened on Sep 23, 2026

• FV-FALCON-HOLD-001 cites a test that cannot fail for the branch it names (mutation-tested)

  pulseengine/relay#483 · Filter by author avrabe opened on Sep 23, 2026
  2

• flying() gates the estimator's attitude reference on hover_thrust — and the shipped component hardcodes 0.5

  pulseengine/relay#481 · Filter by author avrabe opened on Sep 23, 2026

• The 'hover' scenario diverges (74.6 m at 25 s, 202 m at 60 s) — a legacy path superseded by 'flightcore', still shipped as a scenario

  pulseengine/relay#477 · Filter by author avrabe opened on Sep 23, 2026

• SimBackend's accelerometer cannot show thrust — 18 closed-loop hold tests pass on a vehicle that never accelerates

  pulseengine/relay#452 · Filter by author avrabe opened on Sep 17, 2026

• The starvation check keys on RUN status — a matrix run whose jobs are all being picked up alarms as 'queued 30+ min' (#447 was a false positive)

  pulseengine/relay#449 · Filter by author avrabe opened on Sep 17, 2026
  1

• The fleet monitor has never measured starvation: gh isn't installed on the light runners, and every green run since 2026-08-07 was an empty result

  pulseengine/relay#436 · Filter by author avrabe opened on Sep 17, 2026
  1

• MockPhysics applies NO torque — the analytic soak's 12 h hold is 1-D altitude on a vehicle that cannot tilt, and ENDURANCE-P01 draws a controller conclusion from it

  pulseengine/relay#435 · Filter by author avrabe opened on Sep 16, 2026
  1

• FlightCore's hold loop is UNSTABLE when position fixes arrive at ≤10 Hz — zero noise, deterministic, reproduces on SimBackend in <1 s (likely mechanism of #403)

  pulseengine/relay#434 · Filter by author avrabe opened on Sep 16, 2026
  4

• CI jobs wedge in_progress for 30x their normal duration, and the fleet monitor cannot see it (watches queued, not stuck)

  pulseengine/relay#429 · Filter by author avrabe opened on Sep 16, 2026
  5

• docs/PX4-PARITY-ASSESSMENT.md is 82 versions stale and credits unreachable capability as delivered — it answers the wrong way on the exact question it exists for

  pulseengine/relay#422 · Filter by author avrabe opened on Sep 16, 2026

• Documentation integrity: the #388 correction never reached any document that tells someone what to DEPLOY

  pulseengine/relay#419 · Filter by author avrabe opened on Sep 16, 2026

• verus.yml PR trigger is path-filtered on LEAN paths — the Verus gate never runs on a PR that touches a Verus proof

  pulseengine/relay#418 · Filter by author avrabe opened on Sep 16, 2026

• All three required gates can pass vacuously for a PR that touches only the shipped wasm components

  pulseengine/relay#417 · Filter by author avrabe opened on Sep 16, 2026

• Verification-artifact integrity sweep: 15 of 293 cite something that does not resolve; 6 verified artifacts execute nothing in CI

  pulseengine/relay#415 · Filter by author avrabe opened on Sep 16, 2026

• FlightSupervisor — every failsafe, the mode machine and all 19 pre-arm rows — is absent from the shipped component

  pulseengine/relay#414 · Filter by author avrabe opened on Sep 16, 2026

• Published OCI components advertise "Formally-verified geometric SO(3)" while wrapping the unverified legacy PID controller

  pulseengine/relay#412 · Filter by author avrabe opened on Sep 16, 2026
  1

• #388 is incomplete: the legacy 5-stage cascade still lives in wasm/cm/cascade/ and ships cosign-signed in v1.138.0

  pulseengine/relay#411 · Filter by author avrabe opened on Sep 16, 2026
  1

• The required verification gate has no main backstop, and its scope filter is blind to wasm/ — the shipped components

  pulseengine/relay#410 · Filter by author avrabe opened on Sep 16, 2026

• bazel: //:falcon-cascade-coverage fails to fuse — duplicate export cabi_realloc$2 — and is in no workflow

  pulseengine/relay#407 · Filter by author avrabe opened on Sep 16, 2026

• Verus track has verified NOTHING since 2026-09-11 — toolchain cannot find core/std, all 19 proofs fail in 0.1s

  pulseengine/relay#405 · Filter by author avrabe opened on Sep 15, 2026