> Data revision: baf7c2f4307d3a989a8fb9363df8000fda3d9310
> Source files:
- [FINDINGS.md](https://github.com/KogenAI/kogen-bench/blob/baf7c2f4307d3a989a8fb9363df8000fda3d9310/FINDINGS.md)
- [site/content/finding-pages.json](https://github.com/KogenAI/kogen-bench/blob/baf7c2f4307d3a989a8fb9363df8000fda3d9310/site/content/finding-pages.json)

# Findings

## Published results

Every published result has its own page, grouped by comparison, newest first.

### [kogen-loop vs Codex](/comparisons/loop-vs-codex/)

- [A shell that hands control back during long waits: no clear effect on hard tasks](/findings/shell-yield-eu2/) (Unresolved, 31 October 2026)
- [Adding an 'apply_patch' editing tool to kogen-loop did not reduce its output](/findings/apply-patch-edit5/) (No, 31 October 2026)
- [Asking the model for terser text does not change kogen-loop's cost](/findings/verbosity-low-eu6/) (No, 30 October 2026)
- [kogen-loop's connection to the model streams as fast as Codex's](/findings/route-same-host-throughput/) (No, 30 October 2026)
- [kogen-loop costs less than Codex at high effort, on 14 Go tasks](/findings/replayconf-14task-high/) (Yes, 30 October 2026)
- [Report on whether kogen-loop is cheaper, as reliable and faster than Codex: goal not met](/findings/morning-report-20261030/) (Unresolved, 30 October 2026)
- [Did kogen-loop's connection slow down on 30 October? Earlier data, now superseded](/findings/loop-sub-route-throughput/) (Superseded, 30 October 2026)
- [kogen-loop runs fewer shell commands than Codex in old runs](/findings/loop-shell-tail/) (No, 30 October 2026)
- [Sending reasoning back to the model: promising cost, but unresolved on 8 tasks](/findings/loop-high-reasoning-api-replay13/) (Unresolved, 30 October 2026)
- [kogen-loop used about twice Codex's reasoning, for the same total cost](/findings/loop-high-reasoning-api/) (Unresolved, 30 October 2026)
- [On the subscription route kogen-loop used about 40% of Codex's tokens](/findings/h2h30-sub-route-tokens/) (Yes, 30 October 2026)
- [kogen-loop vs Codex on 8 Go tasks: cheaper at medium effort, no clear gap at high](/findings/confirm5-clean-route-8tasks/) (Yes, 30 October 2026)
- [Short test of two kogen-loop 'several steps per request' tools (batch and exec)](/findings/batch-tool-short-test/) (Unresolved, 30 October 2026)
- [A 'batch' tool for kogen-loop packs more work per request but does not cut cost](/findings/batch-tool-confirm-12task/) (No, 30 October 2026)

### [Kogen vs the same agent used alone, on hard tasks](/comparisons/kogen-vs-agents-alone/)

- [Kogen with the shaping fix against Claude Code alone on five Rails tasks (smoke, round 2)](/findings/lab-aor-kogen-c7-round2-20261107/) (Unresolved, 7 November 2026)
- [Correction: three of the nine chosen hard tasks were chosen because of a grader defect](/findings/hk2-hard-set-grading-correction/) (Yes, 1 November 2026)
- [Correction: how the grader affected the five hard-task results](/findings/hard-tasks-grading-correction/) (Yes, 31 October 2026)
- [A newer Kogen driving Codex vs Codex alone: tied on the tasks that could be graded](/findings/hk4-lm-hard-tasks/) (Unresolved, 30 October 2026)
- [Kogen driving Codex vs Codex alone, 6 hard tasks, one MacBook: tied on success](/findings/hk3-mb-hard-tasks/) (Unresolved, 30 October 2026)
- [Kogen driving Codex vs Codex alone, 9 hard tasks, one laptop: tied on success](/findings/hk3-lm-hard-tasks/) (Unresolved, 30 October 2026)
- [Choosing hard tasks for Kogen against Codex alone](/findings/hk2-hard-set/) (Unresolved, 30 October 2026)
- [What Kogen's independent review step did in hard-task runs](/findings/hard-task-review-value/) (Unresolved, 30 October 2026)
- [Kogen driving Claude Code vs Claude Code alone, 9 hard tasks, one laptop](/findings/ccs3-lm-hard-tasks/) (Unresolved, 30 October 2026)
- [Kogen driving Claude Code vs Claude Code alone, 6 hard tasks, one MacBook, second block](/findings/ccs2-mb-hard-tasks/) (Unresolved, 30 October 2026)
- [Kogen driving Claude Code vs Claude Code alone, 6 hard tasks, one MacBook](/findings/cco-mb-hard-tasks/) (Unresolved, 30 October 2026)

### [kogen-loop speed, patches and connection routes](/comparisons/loop-speed-and-routes/)

- [kogen-loop is faster and lighter than Claude Code on the same API key](/findings/lab-cc-layers-b-20261105/) (Yes, 5 November 2026)
- [What keeps the prompt cache warm (unconfirmed probe)](/findings/api-prompt-cache-reuse-probe-20261104/) (Unresolved, 4 November 2026)
- [Message Batches review of landed commits](/findings/api-batch-review-landed-commits-20261104/) (Unresolved, 4 November 2026)
- [A change to Kogen's loop that labels tool errors sends the model exactly the same bytes](/findings/kogen-loop-tool-error-classes-2fc1eb7/) (No, 3 November 2026)
- [Handing small repositories to kogen-loop up front still removes its look-around requests on three servers, with input flat on the Mac](/findings/loop-workspace-snapshot-replication-3hosts/) (Yes, 2 November 2026)
- [kogen-loop handing small repositories to the model up front removes its look-around requests](/findings/loop-workspace-snapshot-166ec7f-eu/) (Yes, 2 November 2026)
- [kogen-loop's latest version passes 13 of 14 tasks; its new cost is shell polling, not failures](/findings/loop-pin-1a3eb18-failed-row-cost-eu2/) (Unresolved, 1 November 2026)
- [kogen-loop's polling and refused-call fixes work: half the polling requests, no refused calls, no task broken](/findings/loop-fixes-9e90bf5-vs-pin-1a3eb18-eu2/) (Yes, 1 November 2026)
- [kogen-loop waiting up to 45 seconds for a shell command removes result-fetching requests on hard tasks](/findings/loop-bounded-shell-wait-166ec7f-eu2/) (Yes, 1 November 2026)
- [Starting tool calls while the model is still streaming does not make kogen-loop faster](/findings/overlap3-ceiling-eu2/) (No, 31 October 2026)
- [One extra instruction sentence made kogen-loop about 17% faster at the same pass rate](/findings/loopnight-instruction-levers/) (Yes, 31 October 2026)
- [What kogen-loop does when the model stream or its shell misbehaves](/findings/loop-stream-shell-fault-fixtures/) (Unresolved, 31 October 2026)
- [The finish-right-away sentence caused no early stops on two hard review tasks, but only one could show it](/findings/l3-stop-at-green-hard-kits/) (Unresolved, 31 October 2026)
- [No faster route to the model was reachable with kogen-loop's own login](/findings/loop-sub-route-honest-probe/) (No, 30 October 2026)
- [Letting kogen-loop's file-writing tool carry a follow-up check: not used by the model](/findings/highwall-write-check/) (No, 30 October 2026)
- [Three speed patches for kogen-loop: none faster or cheaper beyond noise](/findings/api-patch-screen-2nd-host/) (No, 30 October 2026)

### [Test setup, rules and sandboxing](/comparisons/test-setup-and-sandboxing/)

- [Approval delays lack phase attribution](/findings/approval-delay-evidence/) (Unresolved, 8 November 2026)
- [Alarm relays to main: one line per event removes every repeat](/findings/lab-loops-collide-s0-20261105/) (Yes, 5 November 2026)
- [Rule: loop-vs-Codex time and cost are measured only on the two Linux servers](/findings/protocol-split-20261030/) (Yes, 30 October 2026)
- [Test harness change: install dependencies once, retry timed-out approvals](/findings/npm-preflight-cache/) (Yes, 30 October 2026)
- [Why a Kogen approval timed out: the shared 120-second deadline under heavy load](/findings/kogen-approve-deadline-load/) (Yes, 30 October 2026)
- [Four macOS sandbox tools measured against Kogen's old containment needs (historical)](/findings/containment-macos-native-20261015/) (Yes, 15 October 2026)

### [Kogen shaping and delivery-build speed](/comparisons/kogen-shaping-and-builds/)

- [Six of nine Kogen test suites exceed their time budgets on a quiet host](/findings/lab-tests-l4-per-test-profile-20261107/) (Unresolved, 7 November 2026)
- [On macOS, Go tests started detached are scanned by XProtect; attached to ssh they are not](/findings/lab-xprotect-launch-path-20261106/) (Yes, 6 November 2026)
- [The tests request's time budgets cannot yet be priced from recorded logs](/findings/lab-tests-l4-critical-path-20261105/) (Unresolved, 5 November 2026)
- [Unconditional kogen check at landing fails its rule: it refuses good commits and reproduces 1 of 8 breakers](/findings/lab-prelanding-kogen-check-20261105/) (No, 5 November 2026)
- [Reading a Build's final outcome fixes delivery's misreads](/findings/lab-c9-outcome-misread-reader-20261105/) (Yes, 5 November 2026)
- [Batching Kogen's approval step-binding check cuts approval time by over 90%](/findings/build-speed-e1-approve-batch-20261105/) (Yes, 5 November 2026)
- [Retained verification diagnostics and repair locations](/findings/verification-diagnostic-localization-replay-20261103/) (Unresolved, 3 November 2026)
- [Verification timeout diagnostics remain unconfirmed in task runs](/findings/kogen-verification-timeout-evidence-5070052/) (Unresolved, 3 November 2026)
- [Kogen's landed leftover-process sweep still misses a server started by a launcher that closes file descriptors](/findings/kogen-verification-orphan-sweep-6ac87f0/) (No, 3 November 2026)
- [Verification caller cancellation follow-up](/findings/kogen-verification-caller-cancellation-f0884fd/) (Unresolved, 3 November 2026)
- [A rule asking Kogen's shaper for a protected property check saved a valid intent in 1 of 10 pairs against 4 for the adopted build, so we dropped it](/findings/kogen-shaping-row3-verification-separation-0c8852e8/) (No, 3 November 2026)
- [Shaping repair completed paired screen](/findings/kogen-shaping-repair-completed-pairs-2192ae21/) (Unresolved, 3 November 2026)
- [Shaping repair paired screen](/findings/kogen-shaping-repair-before-terminal-submit-2192ae21/) (Unresolved, 3 November 2026)
- [Letting Kogen's shaper decide non-product questions itself looks better on paper, but the change did not cause the gains](/findings/kogen-shaping-decide-non-ux-e572010/) (Unresolved, 3 November 2026)
- [With a stronger model, Kogen's data-stores tool and the adopted build both wrote a valid intent in all 13 pairs, and neither passed a hidden suite](/findings/kogen-shaping-data-stores-tool-13ff132-opus/) (Unresolved, 3 November 2026)
- [Kogen's re-landed data-stores tool wrote fewer valid intents than the adopted build, by the narrowest margin our rule counts](/findings/kogen-shaping-data-stores-tool-13ff132/) (No, 3 November 2026)
- [Telling Kogen's shaper to decide reversible product questions itself did not get one stuck ticket to an Intent](/findings/kogen-shaper-product-decisions-5801b28/) (No, 3 November 2026)
- [Moving Kogen's project setup step into its own component changed none of its code or prompt text](/findings/kogen-setup-initial-e580996/) (No, 3 November 2026)
- [Loop output line boundaries need a triggered confirmation](/findings/kogen-loop-line-cuts-eed48c4/) (Unresolved, 3 November 2026)
- [Three-task Loop evidence wave](/findings/kogen-loop-evidence-wave-three-kits-6e69ffb3/) (Unresolved, 3 November 2026)
- [Executor policy move cannot compile in the adopted assembly](/findings/kogen-executor-phase-api-2f13bd4/) (Unresolved, 3 November 2026)
- [Coordinator validation feedback screen](/findings/kogen-coordinator-lint-routing-e26a4d7/) (Unresolved, 3 November 2026)
- [A readiness regression claim remains tentative after targeted repeats](/findings/kogen-coherent-worseattr-confirmation-20261103/) (Unresolved, 3 November 2026)
- [A bounded follow-up leaves the component update unqualified](/findings/kogen-coherent-set-c92d897-confirmation/) (No, 3 November 2026)
- [A coherent component update remains unqualified](/findings/kogen-coherent-set-c92d897/) (Unresolved, 3 November 2026)
- [A coherent component update regresses task readiness](/findings/kogen-coherent-main-a1a0445/) (No, 3 November 2026)
- [A change to Kogen's review step that scopes its no-defect check made no difference in 2041 retained reviews](/findings/kogen-build-nodefect-scope-269823f/) (No, 3 November 2026)
- [Build changes remain unverified after an incomplete screen](/findings/kogen-build-main-09b177e/) (Unresolved, 3 November 2026)
- [A bug in our Agents-on-Rails task converter would have hidden a correct solution's hidden checks on one new task; no published score is affected](/findings/kogen-aor-kit-evaluate-ids/) (No, 3 November 2026)
- [The adopted Kogen release saved a valid intent in 4 of 12 control runs and passed every hidden check in none](/findings/kogen-aor-control-573d023/) (Unresolved, 3 November 2026)
- [Showing Kogen's builder a small repository's files in its first request cuts its reading steps to zero and build requests by 17%, but passed no more hidden tests (57 of 66 against 58)](/findings/kogen-workspace-snapshot-e6f69a1/) (No, 2 November 2026)
- [Kogen's shaper on Claude Opus 5.5 through Claude Code could not be judged: the test stopped after 47 pairs ran with both arms, too few to tell](/findings/kogen-shaper-opus-claude-code-partial/) (Unresolved, 2 November 2026)
- [Kogen's data-stores change was refused at shaping again on a third machine: 0 of 8 valid intents pooled, against 8 of 8 for the build before it](/findings/kogen-shape-data-stores-816b0b0-st2/) (Unresolved, 2 November 2026)
- [Kogen's data-stores change was refused at shaping in 4 of 4 account-cleanup runs, so its coverage is unmeasured](/findings/kogen-shape-data-stores-816b0b0/) (Unresolved, 2 November 2026)
- [Kogen split into component modules behaved like the build before it in three matched runs](/findings/kogen-extraction-aa-573d023/) (Unresolved, 2 November 2026)
- [Letting Kogen's loop wait 45 seconds for shell commands passes the same hidden tests (55 of 78 each) and saves no requests on these tasks](/findings/kogen-bounded-shell-wait-e6f69a1/) (No, 2 November 2026)
- [Frozen Kogen passed every hidden check in 0 of 18 runs on six Rails tasks, and shaped a valid intent in 6](/findings/kogen-aor-freeze-baseline-3742abe/) (Unresolved, 2 November 2026)
- [Four more runs per task on one machine: current Kogen shaped a valid intent in 3 of 14 counted runs and passed every hidden check in none](/findings/kogen-aor-control-514dab4-st2/) (Unresolved, 2 November 2026)
- [Current Kogen passed every hidden check in 1 of 18 runs on six Rails tasks, and shaped a valid intent in 8](/findings/kogen-aor-control-514dab4/) (Unresolved, 2 November 2026)
- [A second block of Kogen on Claude Opus 5.5 passed every hidden check in 4 of 12 runs on six Rails tasks, and 11 of 12 wrote a valid intent](/findings/kogen-aor-claude-ceiling-514dab4-r2/) (Unresolved, 2 November 2026)
- [With mise's trust setting applied, Kogen on Claude Opus 5.5 passed every hidden check in 1 of 6 runs on a third server, and 3 runs got no valid intent](/findings/kogen-aor-claude-ceiling-514dab4-misetrust-eu/) (Unresolved, 2 November 2026)
- [With mise's trust setting applied, Kogen on Claude Opus 5.5 passed every hidden check in 2 of 12 runs on six Rails tasks, and 7 runs were still refused at shaping](/findings/kogen-aor-claude-ceiling-514dab4-misetrust/) (Unresolved, 2 November 2026)
- [Kogen on Claude Opus 5.5, third server: 3 of 6 runs landed a build and 1 passed every hidden check; pooled 2 of 18](/findings/kogen-aor-claude-ceiling-514dab4-eu/) (Unresolved, 2 November 2026)
- [Kogen on Claude Opus 5.5 passed every hidden check in 1 of 12 runs on six Rails tasks, and every valid intent it shaped landed](/findings/kogen-aor-claude-ceiling-514dab4/) (Unresolved, 2 November 2026)
- [A test-authoring paragraph in Kogen's builder prompt did not shorten builds on four Go tasks](/findings/test-authoring-go-screen/) (No, 1 November 2026)
- [Kogen's new shaping example check refuses good specifications and loses 7 hidden passes in 65 shapes](/findings/shape-example-check-3ea2e23-screen/) (No, 1 November 2026)
- [Turning on the request audit in Kogen shaping lost badly to unpatched Kogen over two Go repetitions and stays dropped](/findings/s3-audit-on-go-pooled/) (No, 1 November 2026)
- [The stop-when-checks-pass sentence in Kogen's builder prompt lost to unpatched Kogen over two Go repetitions](/findings/p5-stop-at-green-go-pooled-eu2/) (No, 1 November 2026)
- [The patched builder's losses were caused by defects written before it started, not by finishing early](/findings/p5-loss-attribution/) (Unresolved, 1 November 2026)
- [A coverage ledger in Kogen shaping did not earn its keep over two Go repetitions and is dropped](/findings/p4-ledger-go-pooled/) (No, 1 November 2026)
- [A coverage-ledger patch for Kogen showed no gain on six hard Rails tasks, where no run passed](/findings/p4-ledger-aor-screen/) (No, 1 November 2026)
- [An unstated-requirement-pass patch for Kogen showed no gain on six hard Rails tasks, where no run passed](/findings/p2-unstated-aor-screen/) (No, 1 November 2026)
- [Updating Kogen's built-in loop to tonight's fixes removes refused shell calls but shows no gain in hidden passes (32 vs 31 of 52)](/findings/loop-repin-9e90bf5-in-kogen-eu2/) (Unresolved, 1 November 2026)
- [High reasoning effort for Kogen's shaper passes no more hidden tests than medium (57 of 78 each)](/findings/kogen-shaper-high-effort-e6f69a1/) (No, 1 November 2026)
- [Kogen's current main passes more hidden tests than the build before its last nine commits (31 vs 23 of 39), mostly because the example check was removed](/findings/kogen-release-e6f69a1-vs-69d09e1/) (Yes, 1 November 2026)
- [Kogen's cold prompt-cache requests are conversation starts, and putting the testing rules first does not warm them](/findings/kogen-prompt-cache-cold-starts/) (No, 1 November 2026)
- [Kogen's newer build matched the previous build on 13 Go tasks over three repetitions](/findings/b1-base-qualification-go-pooled/) (Unresolved, 1 November 2026)
- [Kogen's newer build with the delivered fixes did no worse than the previous build on 13 Go tasks](/findings/b1-base-qualification-go-eu2/) (Unresolved, 1 November 2026)
- [Which of six ways to reuse or skip checks in Kogen builds could save time](/findings/verification-reuse-replay/) (Unresolved, 31 October 2026)
- [Three shaping changes screened: the second-host half was invalid, the valid half shows no gain](/findings/shape-screen-terra-b/) (Unresolved, 31 October 2026)
- [A clearer refusal message for shaping changed nothing: both versions accepted 25 of 26](/findings/shape-screen-s5/) (No, 31 October 2026)
- [Allowing only one resubmission broke Kogen's shaping step; allowing five matched the default](/findings/shape-retry-levers-k11/) (No, 31 October 2026)
- [A cheaper model at medium effort could not get Kogen's shaping step accepted (7 of 78)](/findings/shape-luna-medium-refusals/) (No, 31 October 2026)
- [Switching on Kogen's request audit lost four net tasks on 13 Go tasks, so the patch was dropped](/findings/s3-audit-on-go-screen/) (No, 31 October 2026)
- [A stop-when-checks-pass sentence in Kogen's builder prompt held the hidden result and cut effort on the first 13 Go tasks](/findings/p5-stop-at-green-go-screen/) (Yes, 31 October 2026)
- [A coverage ledger in Kogen shaping did not show a gain on the first 13 Go tasks](/findings/p4-ledger-go-screen/) (Unresolved, 31 October 2026)
- [Adding review probes to Kogen lost two net tasks on 13 Go tasks, so the patch was dropped](/findings/p3-review-probes-aor/) (No, 31 October 2026)
- [An unstated-requirement pass in Kogen shaping did not show a gain on the first 13 Go tasks](/findings/p2-unstated-go-screen/) (Unresolved, 31 October 2026)
- [An unstated-requirement pass in Kogen shaping lost to unpatched Kogen over two Go repetitions and is dropped](/findings/p2-unstated-go-pooled/) (No, 31 October 2026)
- [Where the time goes in Kogen builds and shaping sessions](/findings/delivery-path-timing/) (Unresolved, 31 October 2026)
- [A prototype that skips an already-passed starting check: safe, but small headroom](/findings/baseline-check-reuse-prototype/) (Unresolved, 31 October 2026)

### [Kogen shaping component](/comparisons/kogen-shaping/)

- [Setup evidence in contract review remains unproven](/findings/shape-protected-setup-evidence-screen/) (Unresolved, 8 November 2026)
- [Historical and current shaping releases](/findings/lab-shaping-historical-current-20261108/) (Unresolved, 8 November 2026)
- [Screen of shaping evidence and probe handling](/findings/lab-shaping-evidence-probe-screen-20261108/) (No, 8 November 2026)
- [Evidence-path guidance before and after rejection](/findings/lab-shaping-dual-path-guidance-screen-20261108/) (No, 8 November 2026)
- [Decision check and shaping readiness](/findings/lab-shaping-decision-bisect-20261108/) (Unresolved, 8 November 2026)
- [Combined evidence and transient-output guidance](/findings/lab-shaping-combined-guidance-screen-20261108/) (No, 8 November 2026)
- [Guidance for carrying shaping decisions](/findings/lab-shaping-carried-decision-screen-20261108/) (No, 8 November 2026)
- [Variation in shaping readiness](/findings/lab-shaping-baseline-noise2-20261108/) (Unresolved, 8 November 2026)
- [Shaping guidance for assumptions and evidence](/findings/lab-shaping-assumption-guidance-screen-20261108/) (No, 8 November 2026)
- [Existing-behavior evidence in shaping](/findings/lab-shaping-observed-evidence-screen-20261107/) (No, 7 November 2026)
- [Initial evidence-path instruction in shaping](/findings/lab-shaping-initial-evidence-screen-20261107/) (No, 7 November 2026)
- [Evidence-path instructions in shaping](/findings/lab-shaping-evidence-path-screen-20261107/) (No, 7 November 2026)
- [Deciding instead of asking shortens shaping's question loop but does not produce Intents](/findings/lab-shaping-converge-20261107/) (No, 7 November 2026)
- [Shaping readiness on a fixed request set](/findings/lab-shaping-climb-round0-20261107/) (Unresolved, 7 November 2026)
- [Preventing transient bytecode in shaping checks](/findings/lab-shaping-bytecode-screen-20261107/) (No, 7 November 2026)
- [Variation in shaping readiness](/findings/lab-shaping-baseline-noise1-20261108/) (Unresolved, 7 November 2026)
- [Naming the decision line and controls.json gets Rails shapes through shaping](/findings/lab-shaping-aor-diagnosis-20261107/) (Yes, 7 November 2026)
- [Naming the gate forms in the shaper prompt gets small requests through shaping; the Rails tasks stop at later gates](/findings/lab-shaping-small-requests-20261107/) (No, 6 November 2026)
- [The review-size bound counts whole files the review never carries, and the one change that passes it removes the gate](/findings/lab-rc-c2-scope-bound-before-readiness-20261106/) (No, 6 November 2026)
- [The decision-record gate costs one shaping attempt, but other gates end these shapes before readiness](/findings/lab-rc-c-decision-gate-before-readiness-20261106/) (No, 6 November 2026)
- [The second shaper text changes leave all six cells at the 30-minute deadline, with the first draft using most of it](/findings/lab-aor-shaping-prompt-screen2-20261106/) (Unresolved, 6 November 2026)
- [The shaper prompt change gets one of five refused tasks to readiness; the rest stop at prose statements and later gates](/findings/lab-aor-shaping-prompt-screen-20261106/) (No, 6 November 2026)
- [Request-only shaping refusals are mostly prompt-versus-gate contradictions, and gate-side changes alone turn 2 of 5](/findings/lab-aor-shaping-no-answerer-gates-20261106/) (No, 6 November 2026)
- [Handoff questions stopped shaping before the question check, not after](/findings/lab-shape-handoff-question-s0-20261105/) (No, 5 November 2026)
- [A model-free scope inventory names every owner-keyed store the hidden checks touch; the counted shaper cells were not run](/findings/lab-k-shape-scope-enumeration-s1-20261105/) (Unresolved, 5 November 2026)
- [Static checks on shaped files stop 1 of 22 defects the builder refuses](/findings/lab-c2-static-checks-20261105/) (No, 5 November 2026)
- [Shaping recovery on Claude: one rescue, otherwise a tie (tentative)](/findings/kogen-shaping-recovery-claude-20261105/) (Unresolved, 5 November 2026)
- [Shaping recovery on Claude: second smoke run, unresolved](/findings/kogen-shaping-claude-route-confirm-c3-20261105/) (Unresolved, 5 November 2026)
- [Bounded shaping recovery confirmation](/findings/kogen-shaping-recovery-confirmation-stop-41d70052/) (Unresolved, 4 November 2026)
- [Repair after shaping refusals remains tentative](/findings/kogen-shaping-bounded-refusal-recovery-41d70052/) (Unresolved, 4 November 2026)

### [Kogen build component](/comparisons/kogen-build/)

- [A behaviour projection leaves every recorded false red of the budget-off test outside it](/findings/lab-tests-l2-projection-20261105/) (Yes, 5 November 2026)
- [Slow checks re-run on an unchanged tree: little to save, and already shipped](/findings/lab-slow-checks-rerun-s0-20261105/) (No, 5 November 2026)
- [Changed-first checks in build rounds: the targeted test run is not cheap enough](/findings/lab-k-round-full-suite-s1-20261105/) (No, 5 November 2026)
- [An allow-list for the budget-off consumer test: not measured](/findings/lab-c6-consumer-overcoupling-allowlist-20261105/) (Unresolved, 5 November 2026)
- [Build component check on Claude stays unresolved](/findings/kogen-build-claude-route-confirm-c3-20261105/) (Unresolved, 5 November 2026)
- [Extracting one release file cuts the biggest MacBook test's temp files by 46%](/findings/build-speed-e6d-assembly-single-member-extract-20261105/) (Yes, 5 November 2026)
- [Sharing a source copy does not cut MacBook temp-file load](/findings/build-speed-e6-fsevents-temp-trees-20261105/) (No, 5 November 2026)
- [Verify timeouts and land-red under load stay unresolved](/findings/build-speed-e3-verify-load-20261105/) (Unresolved, 5 November 2026)
- [Shared Go lock for small suites stays unresolved](/findings/build-speed-e2-go-lock-20261105/) (Unresolved, 5 November 2026)
- [Build component head with a fixed Intent](/findings/kogen-build-mainhead-supplied-intent-20261104/) (Unresolved, 4 November 2026)
- [Build main head on Claude: no clear difference (tentative)](/findings/kogen-build-claude-supplied-intent-20261104/) (Unresolved, 4 November 2026)
- [Build head paired screen remains tentative](/findings/build-release-mainhead-20261104/) (Unresolved, 4 November 2026)

### [Kogen review component](/comparisons/kogen-review/)

- [Why two review verdicts were refused](/findings/lab-review-invalid-quote-and-unread/) (Unresolved, 8 November 2026)
- [A refused review submission should not make its requirement texts binding](/findings/lab-rc-b-review-listed-requirements-20261106/) (Yes, 6 November 2026)
- [No review arm of the wide-change Lab passes Stage 0; today's review already cuts 9 parts to 3](/findings/lab-k-review-wide-stage0-20261105/) (No, 5 November 2026)
- [Review component check on Claude stays unresolved](/findings/kogen-review-claude-route-confirm-c3-20261105/) (Unresolved, 5 November 2026)
- [Review component head with a fixed Intent](/findings/kogen-review-mainhead-supplied-intent-20261104/) (Unresolved, 4 November 2026)
- [Review head regression not confirmed](/findings/kogen-review-mainhead-confirm-20261104/) (No, 4 November 2026)
- [Review component on Claude: no difference attributable (tentative)](/findings/kogen-review-claude-supplied-intent-20261104/) (Unresolved, 4 November 2026)
- [Saved-candidate reviews after four changes](/findings/review-climb-saved-candidates-20261103/) (Unresolved, 3 November 2026)

### [Kogen verification component](/comparisons/kogen-verification/)

- [A shaper's probe and the readiness run used different environments](/findings/lab-probe-env-shaper-vs-readiness-20261107/) (Yes, 7 November 2026)
- [Archiving per-Intent step files leaves the spec trace unchanged, but many of them already name tests that no longer exist](/findings/lab-tests-l8-step-files-20261105/) (No, 5 November 2026)
- [The replay's Quint batch is fast, but its generator rejects 7 of 17 valid acceptance runs](/findings/lab-tests-l5-quint-replay-20261105/) (No, 5 November 2026)
- [On Linux the wall-time deadline test does not fail; its pre-setup work is 4x shorter than on the Mac](/findings/lab-rc-l1-walltime-linux-20261105/) (Yes, 5 November 2026)
- [The wall-time deadline test starts its clock too early; the loaded run did not happen](/findings/lab-rc-l1-walltime-20261105/) (Yes, 5 November 2026)
- [Isolated, parallel-safe generated acceptance: no detector with a payoff passes](/findings/lab-k-verify-isolated-s1-20261105/) (No, 5 November 2026)
- [A probe copy with git history lets git-reading observers pass](/findings/lab-c10-probe-git-copy-20261105/) (Yes, 5 November 2026)
- [Reaping while the check runs ends the Linux helper's hangs, with no orphans](/findings/lab-c1-reaper-confirm-20261105/) (Yes, 5 November 2026)
- [Linux check helper leaves zombies; reaping while the check runs fixes it in a first run](/findings/lab-c1-linux-reaper-zombies-20261105/) (Unresolved, 5 November 2026)

### [Kogen Context component](/comparisons/kogen-context/)

- [A new-nonce instruction detected 8 of 8 forced compactions in a small Claude Code test](/findings/lab-compaction-p3-newnonce-20261105/) (Yes, 5 November 2026)
- [Context packet shows no time-to-patch gain in a screen](/findings/kontext-matched-2-verified-patch-time-20261105/) (Unresolved, 5 November 2026)
- [Packet-only localization on 11 more cases](/findings/kontext-api-localization-matched2-20261105/) (No, 5 November 2026)
- [Detecting compaction without a model: a transcript watcher always works, token drop only on long sessions](/findings/compaction-watcher-token-drop-p3a-20261105/) (Yes, 5 November 2026)
- [A brief marker did not detect compaction; the hook did](/findings/brief-marker-compaction-p1-20261105/) (No, 5 November 2026)
- [Hook-free compaction detection: markers fail, a self-check works, the hook needs the launcher](/findings/brief-compaction-detection-p2-20261105/) (Unresolved, 5 November 2026)
- [Packet-only localization over the API](/findings/kontext-packet-only-localization-api-20261104/) (Unresolved, 4 November 2026)
- [Context packet did not speed up four tasks](/findings/kogen-context-failure-packet-screen-20261104/) (Unresolved, 4 November 2026)

## Source summary by family

The family-by-family source summary is reproduced below. Figures that are not reconciled to a claim ledger remain source-reported; follow the source and related claim records for the limits.

- [F01 — Shape, Intent and plan](/findings/f01/)
- [F02 — Isolation, landing and interface](/findings/f02/)
- [F03 — Harness and loop](/findings/f03/)
- [F04 — Models, efforts and routing](/findings/f04/)
- [F05 — Context, Jev and cache](/findings/f05/)
- [F06 — Checks, review and recovery](/findings/f06/)
- [F07 — Tasks, grading and method](/findings/f07/)
- [F08 — Language and executable spec](/findings/f08/)
- [F09 — Campfire and web stacks](/findings/f09/)
- [F10 — Agent-facing output](/findings/f10/)
- [F11 — Claimed cross-cutting results](/findings/f11/)
- [F12 — Explicit lever and recipe arms](/findings/f12/)
- [F13 — Late-September product mechanisms](/findings/f13/)
- [F14 — Context-store and task policy](/findings/f14/)
