Skip to content

Independent reassessment of all prior findings

The old CSV has 51 entries, not 55: reserved 029 and 043–045 were merged. This table covers every actual row. “Fixed” is scoped to the original failure and cited mechanism, not certification of an entire subsystem. No prior private exploit evidence or live repository settings were assumed available.

Prior ID Previous reported state Independent disposition and evidence
ICX-AUD-001 fixed Partially verified remediation; broader latency claims require scale tests: Source persists HDC snapshots and ships daemon. This component benchmark does not prove 10k real-binary hook p99.
ICX-AUD-002 fixed Confirmed fixed for labeling; fixture-dependent effectiveness remains unverified: CLI/docs now call the number byte proxy; provider response-token measurements not performed.
ICX-AUD-003 fixed Confirmed fixed for identified placement/stripping defects: Content-block insertion, preservation/count of client breakpoints and TTL ordering logic; original broad prefix-rewrite allegation was overstated. Provider hit effectiveness remains unverified.
ICX-AUD-004 fixed Confirmed fixed for inspected event fields: Shape-aware Bash object, PreCompact stdout, proper stop input; source and public docs. Every generic tool schema has not been exercised live.
ICX-AUD-005 partial Partially fixed: Built-in duplicate context removed and configuration docs corrected; unsupported MCP replacement persists (035), rollout zstd/CLI wiring absent.
ICX-AUD-006 partial Partially fixed: Invalid ready-to-run shim claims removed, absence of Python integration disclosed; no shim ships.
ICX-AUD-007 fixed Partially fixed: Stop/budget override repaired, named config locations corrected; undocumented UI-hidden behavior remains unverified.
ICX-AUD-008 fixed Confirmed fixed for original version-negotiation defect; new envelope validation issue 031 is separate
Partially fixed: Legacy negotiation and modern discovery added; malformed input validation still incomplete (031).
ICX-AUD-009 fixed Confirmed fixed: PreToolUse no longer echoes output-only permission fields for Claude/Codex.
ICX-AUD-010 fixed Confirmed fixed for default CLI trust path: workspace::resolve, prepare, verify, owner/store seals; repository .icx is not selected implicitly. Existing trust tests exercise replacement and modified rows.
ICX-AUD-011 fixed Confirmed fixed for prefix-marker bypass: Exact emitted-block digest membership plus Tier-0 escaping replaces prefix-based trust. This cannot prevent natural-language prompt injection; that limitation is documented.
ICX-AUD-012 fixed Confirmed fixed for default CLI route: Token check on all routes, loopback bind restriction in CLI; opt-out flags explicitly weaken controls.
ICX-AUD-013 fixed Confirmed fixed: Private directory/file creation and .gitignore rewrite; adoption tightens owned regular files.
ICX-AUD-014 fixed Confirmed fixed for original unbounded stdin defect; elapsed-time/concurrency limits remain separate hardening
Partially fixed / bounded: CLI hook/MCP bounds and fold depth exist. Single-request limits are not a complete concurrency or elapsed-time budget.
ICX-AUD-015 fixed Partially fixed: URL/vendor/next-line fixtures added; YAML block scalars fail (001)
ICX-AUD-016 fixed Confirmed fixed for reported inline-PEM bug: Inline spans replace start..end and continue; runaway bounds remain documented
ICX-AUD-017 fixed Partially fixed: Current threat model now names implemented controls and residual risks. Its blanket guarantees require the qualifications above and the bus/integration reports.
ICX-AUD-018 fixed Confirmed fixed in source/tests: Nonblocking no-follow transcript open plus regular-file check.
ICX-AUD-019 fixed Confirmed fixed: File descriptor flags, fstat ownership/regular-file/mode checks are present. Metadata/path race freedom against same-user writers is outside scope.
ICX-AUD-020 fixed Partially fixed: Private scratch files, RAII cleanup and periodic/startup sweeps exist. SIGKILL residue persists until a sweep as documented. The additional continuous-stream cancellation lead 034 was rejected after real-curl countertesting.
ICX-AUD-021 fixed Confirmed fixed for secret-free full-byte storage: Independent four-fixture equality probe plus separate storage redact/presentation escaping paths; redaction itself is intentionally lossy.
ICX-AUD-022 fixed Partially fixed: MMU checked-to-unchecked fallback removed; documented unchecked/truncating public API remains.
ICX-AUD-023 fixed Confirmed fixed for silent overwrite: DAG collision guard and checked CAS insertion reject divergent digest/node identity. No practical 64-bit collision search was attempted.
ICX-AUD-024 fixed Confirmed fixed: Hex epoch formatting, 4–8 digit parser and DAG round-trip test
ICX-AUD-025 fixed Confirmed fixed at slice_hint: Invalid hints not rendered and rejected on parse; Profile A skeleton separate defect 002
ICX-AUD-026 fixed Confirmed fixed for parser input: Iterative max 64 materialized links; no claim on arbitrary manually built/serde-recursive trees
ICX-AUD-027 fixed Confirmed fixed in source: Spread high-bit masks and non-single-chunk fixture exist. Independent repository-wide deduplication benchmark was not repeated here.
Confirmed fixed for varied CDC fixture: Current CAS fixture uses deterministic varied lines and reports chunk count.
ICX-AUD-028 fixed Substantially fixed, remaining extrapolation caveat: Native flags recorded, portable/native repeats measured here; ICX-CODEX-072.
ICX-AUD-030 fixed Confirmed fixed for stale Chapter 08 struct: PDF now has current fields and six-way split
ICX-AUD-031 fixed Partially fixed: Q1.14/Q15 mismatch 005; current core spec improved but AHTL chapters remain stale
ICX-AUD-032 partial Partially fixed via supplement: Historical exactness review unchanged; corrections file supplies arithmetic and provenance caveats
ICX-AUD-033 fixed Partially fixed: Dated corrections identify ACM/LCM/RVQ attribution issues; current accepted ADR-0003 still repeats misattributed cost/TTFT and provider-hit assertions (100). All 22 arXiv IDs resolve, not every paper claim reproduced.
ICX-AUD-034 fixed Confirmed fixed for explicit Cargo/REUSE crate declarations; no legal certification: Adapters/proxy manifests now specify GPL AND AGPL; first-party root GPL, notices and license texts inspected. Combined-work interpretation remains counsel scope.
ICX-AUD-035 deferred Open/deferred: ICX-CODEX-074; same unresolved terms persist.
ICX-AUD-036 partial Partially fixed: REUSE now separately labels Zep NC-SA and remaining arXiv PDFs; non-exclusive papers are link-only in index. Unicode issue persists; no complete license attestation for every embedded image.
ICX-AUD-037 partial Partially fixed / live settings unverifiable: SECURITY.md now provides private reporting URL and response target. Required-check and signatures-branch protection not inferred from YAML.
ICX-AUD-038 partial Partially fixed: Asset input directory is configurable; patents ignored in tracked rules; history and binary duplication remain. No claim all historical personal paths are purged.
ICX-AUD-039 partial Partially fixed: Properties/fuzz targets exist; Profile A hostile fields and YAML scalar syntax absent from shown tests
Partially fixed: Property tests/fuzz exist; new mutation survivors remain, ICX-CODEX-073.
ICX-AUD-040 fixed Confirmed fixed for original SHA/toolchain/policy issues; release CI-success gap 070 is separate
Pins and deny policy fixed; no complete CI success predicate in release: Supply-chain report ICX-CODEX-070.
Confirmed fixed for enumerated pins/policy; doctor partial: Full SHA pins, exact toolchain/tools and deny policy verified. Doctor on uninitialized CI state remains shallow; release ancestry defect newly identified.
ICX-AUD-041 fixed Confirmed fixed for reported unused dependencies: Old proxy codec/serde removed; CLI serde is now Unix-only and used by daemon. cargo-machete result in gate evidence.
ICX-AUD-042 fixed Confirmed fixed for duplicate results: Ranked hits dedup by ArtifactId; candidate-crowding lead remains unverified
Confirmed fixed in source: Tail candidates deduplicate artifact IDs; scale/recall quality belongs to report 02.
ICX-AUD-046 fixed Partially fixed: Main bus documentation now describes file I/O and writer locks accurately; stale module/API titles and diagram/ADR language survive (see report 08).
ICX-AUD-047 fixed Partially fixed: Single transaction repairs multi-file atomicity; sealed MAC repairs untrusted derived state for CLI. Unsealed/legacy DAG load remains less validated; daemon acknowledgment durability is ICX-CODEX-030.
ICX-AUD-048 fixed Unverifiable in graph alone: Adapter clock mode determines timestamps; graph itself accepts caller times
Confirmed fixed as documented opt-in: use_wall_clock and monotonic clock implementation; deterministic library default remains intentional.
ICX-AUD-049 fixed Partially fixed: Repetitive success, varied-corpus collapse 004
Separate mathematical/retrieval report owns reassessment: Component target query is planted and cannot alone demonstrate general recall.
ICX-AUD-050 fixed Partially fixed: i32 removes demonstrated i16 saturation; encoding CPU profile not independently refreshed
Source remediation visible; measured implementation parity is separate: No assertion that one microbenchmark proves saturation behavior.
ICX-AUD-051 fixed Partially fixed in prose: Current subsystem spec says inspired; historical and AHTL prose still contains stronger claims
ICX-AUD-052 open Open limitation, partly mitigated: Inverse check rejects opposite orthants, cannot prove original-event suffix identity
ICX-AUD-053 fixed Confirmed fixed as disclosure: hv64 linear collision warning present; no claim hash itself became collision resistant
ICX-AUD-054 fixed Partially fixed: Typed publish failures and loss counters exist; expected sequence invariant still missing (ICX-CODEX-032). The MMU uses ordinary workspace-filtered polling, not a complete loss-repair protocol.
ICX-AUD-055 fixed Partially fixed: Existing test uses two handles in one process. This audit adds a real two-process measurement; it does not demonstrate all four real harnesses or establish their p99 propagation bound.
Partially verified: Two-handle same-process test is distinct from four OS processes/agents; no four-agent timing established here.

No new issue is labeled regressed without comparative old-revision reproduction. Current counterexamples can instead demonstrate a residual or broader coverage gap.