fix(sparq-zk): Implement the noir#13314 constraint-level guard specified in research/noir-bounded-ve

[sparq-orchestrator]

🤖 SPARQ agent

What / why

Automated implementation of the trusted task in #5668, routed to sparq-reviewer on
Claude Opus 5 (alias opus5, provider model claude-opus-5).

Fixes #5668

Local gate

• Policy profile: crate-scoped
• Result: passed before push

Merge posture

DRAFT — pending cross-provider review. Publish never arms; arming happens ONLY in the registry
review-fix approve path (arm_auto_merge=true), gated on an opposite-provider approve
verdict with ci-summary / gate as the objective backstop.

[sparq-orchestrator]

🤖 SPARQ agent — cross-provider review round 1 recorded.

[sparq-orchestrator]

🤖 SPARQ agent — cross-provider review round 1: approve.

The documentation-only change is internally consistent and appropriately conservative for a ZK soundness-sensitive surface. It clearly states that no toolchain execution occurred, does not claim the capacity guard is proven or externally audited, distinguishes a serialization-compatibility risk from a verified incompatibility, and preserves the existing non-vacuity matrix. The repository context supports the concrete local claims about the pinned nargo/bb versions, the ignored-test CI command, and the currently locked dependency majors. The proposed round-trip pre-flight is a reasonable prerequisite to prevent serialization failure from masquerading as a meaningful forged-witness rejection. No executable behavior or test is added, so there is no new test-path validity defect to block this record update.

[sparq-orchestrator]

🤖 SPARQ agent

Arming on cross-provider approve. Trust-surface audit trail (complete): (live security label: routing match_labels / trust:* posture recomputed at arm time) @ b33e375. Post-merge review welcome; revert-and-reopen is the escalation path.

[github-actions]

🤖 SPARQ agent — autonomous orchestration.

This PR was dequeued from the merge queue (reason: CI_TIMEOUT). Its auto-merge arm has been disabled and review:pass swapped for review:changes, so the fix pipeline admits it on the next enumeration tick (durable feedback loop for registry#556).

The failed merge-group run was not discoverable from the Actions API at notification time (no non-success merge_group run recorded for gh-readonly-queue/*/pr-5714-*) — check this PR's checks tab for the queue history.

Reason CI_TIMEOUT is queue/infra-flavoured (likely not a fault in this PR's diff); it is still routed to review:changes because the fix lane's ci-flavour owns re-runs.

[sparq-orchestrator]

🤖 SPARQ agent — fix round 1 executed by opus5.

[sparq-orchestrator]

🤖 SPARQ agent — recorded nochange for review round 1.

[sparq-orchestrator]

🤖 SPARQ agent — fix round 1 executed by opus5.

[sparq-orchestrator]

🤖 SPARQ agent — recorded nochange for review round 1.

[sparq-orchestrator]

🤖 SPARQ agent — the autonomous review loop parked this PR: two consecutive fix attempts made no change (fixer judges the findings spurious)

This is the MACHINE-owned capacity park (review:parked), not a human question: it remains a DRAFT and will not be auto-armed. A human can re-admit it by removing the live machine park label(s) — review:parked here and status:parked on the source issue (whichever are present; a needs:user unlabel on either surface also opens the budget window) — the budget restarts from the latest gesture.

[github-actions]

🤖 SPARQ agent — autonomous orchestration.

This PR was dequeued from the merge queue (reason: CI_TIMEOUT). Its auto-merge arm has been disabled and review:pass swapped for review:changes, so the fix pipeline admits it on the next enumeration tick (durable feedback loop for registry#556).

Failed merge-group run(s) for this PR's queue branch:

• ci-summary (failure)

Reason CI_TIMEOUT is queue/infra-flavoured (likely not a fault in this PR's diff); it is still routed to review:changes because the fix lane's ci-flavour owns re-runs.

[sparq-orchestrator]

🤖 SPARQ agent — automatically re-admitted this MACHINE capacity park: its own starvation cause can no longer be observed, and the fleet is demonstrably healthy.

This park is older than the rolling model-health window, so whether the specific condition that parked it has cleared is NOT provable any more — leaving it would make an automatic hold a permanent one. Instead the fleet has recorded sustained successful runs across multiple accounts with no launch failure, the most recent at 2026-08-02T22:14:52Z (evidence fleet-health/anthropic/f79434c3ffba014f/30769217549.1 — provider/account-fingerprint/run from the model-health window; no raw handle). That is a HEURISTIC about fleet health, not proof that this park's own cause cleared. The machine park label(s) are being removed and the review loop re-admitted with a real budget window.

This consumes that evidence EXACTLY ONCE: the same evidence can never re-admit this PR again, a later park needs FRESH evidence that has not been consumed (for the cause-recovery route, a new outage-and-recovery pair), and at most 2 automatic re-admissions are ever granted to one PR — past that the loop stops and asks a human. A human hold (review:needs-user / needs:user) is never touched by this path, and neither is a park a human applied by stamping one of those human-owned labels. A park a human applied by stamping the MACHINE-owned soft hold is re-admitted here ONLY when this bot's OWN park-reason receipt already classified the episode class=capacity — the machine never clears a park it never classified. To place a hold no machine may lift, use review:needs-user / needs:user.

[jeswr]

Local ci-fast gate (GitHub Actions runners are down; Jesse approved merging on local runs at 20:19 UTC). The PR head was merged locally with main a496ec9 (not pushed), and every step of .github/workflows/ci-fast.yml was run: clippy on the core crates passed; nextest passed (3009 run, 3009 passed, 27 skipped); doctests passed; W3C conformance scored 1225 pass + 4 documented divergences = 1229 (ratchet 1229). rustfmt is non-blocking in ci-fast and shows differences, as it does on main.

Benchmark: not applicable. The change is documentation only; no code path changes.

---

Generated by Claude Code