diff --git a/doc/plans/2026-10-26-chat-catalog-eval-goal.md b/doc/plans/2026-10-26-chat-catalog-eval-goal.md
index 2e501829d88..0c62ec0614f 100644
--- a/doc/plans/2026-10-26-chat-catalog-eval-goal.md
+++ b/doc/plans/2026-10-26-chat-catalog-eval-goal.md
@@ -2,6 +2,16 @@
Date: 2026-10-26
+## Authoritative user steering — 2026-10-27
+
+All new local eval admissions are stopped. Already-active local children may finish their existing work and cleanup; no further local initial, final-B, infrastructure-health or provider qualification launches are authorized. Preserve the complete local measurement/ordinal histories and original evidence as a historical lane.
+
+All future eval execution must run in GitHub Actions on the existing AWS spot runners, using the new Paperclip runner **OpenCode** provider through **OpenRouter**, model **`deepseek/deepseek-v4.1-flash-1017`**. The remote campaign requires a separately named execution/product/database/model identity and separately reported rate across all244 scenarios. Do not carry local passes into the remote rate, and do not execute obsolete local source merely to preserve the previous queue.
+
+Carry the audited coherent harness corrections into the remote source. Preserve prior ordinal histories. The manager approved one new remote initial measurement for every scenario and at most one coherent correction measurement per scenario. The representative remote smoke counts as its scenario’s remote initial. Preserve all prior local ordinals separately; this is not an unlimited repair or unchanged rerun policy. Production/core fixes remain isolated and unmerged unless separately authorized.
+
+Immediate deliverables are an inspectable safe results page, an exact closed local checkpoint, and evidence-based accounting of elapsed wall time, stages reached, harness/product/model/prerequisite outcomes, actual usage and operational duplicate/reconciliation history. The original goal below remains historical context where superseded by this section.
+
> Execution state: This file preserves the requested goal; it is not completion evidence. Before execution, reconcile the actual state of PR #17794 rather than assuming it merged.
Turn the complete Paperclip chat-scenario catalog into executable Product E2E evals, run every scenario, give each failing scenario one fix-and-rerun cycle, and ship according to the rules below.
diff --git a/doc/plans/2026-10-26-chat-catalog-eval-progress.md b/doc/plans/2026-10-26-chat-catalog-eval-progress.md
index 11c0547b795..4a6ff3641ce 100644
--- a/doc/plans/2026-10-26-chat-catalog-eval-progress.md
+++ b/doc/plans/2026-10-26-chat-catalog-eval-progress.md
@@ -4,17 +4,20 @@ Updated: 2026-10-27. This is the single progress record for [the saved goal](202
## Current state
+- **User steering supersedes the local queue:**all new local admissions are stopped. All admitted children have finished and exact owned absence is verified; no remaining76/22 local initials, local B or new health launches. Future execution is GitHub Actions on existing AWS spot runners, using the new Paperclip runner OpenCode provider via OpenRouter model `deepseek/deepseek-v4.1-flash-1017`. Local histories remain preserved; their passes cannot qualify the separately named remote lane. The manager approved a new remote initial for every scenario plus at most one coherent correction measurement, with distinct history and no erased local ordinals.
+- **Readable results:**`tests/runner-e2e/results/chat-catalog-campaign/interim-results.html` now shows all244 searchable outcomes plus frozen public canonical query/context/intended outcome, conservative stage proofs, edited diagnoses, correction commits, evidence references, elapsed gaps and six open/unmerged core PRs. The local checkpoint is closed. Proofs establish105 saved canonical queries,102 exact canonical-native admissions,14 returned feature grades and110 observation-grade snapshots; unproven stages stay unknown. Eighteen measurements have confirmed zero native IDs, two lack bound run-count proof; zero native alone is not a setup classification. Recorded reservation span is15.539h; union of completed attempt wall intervals4.719h (not model time), sum5.879h (overlaps counted). Two large no-measurement gaps total8h26m and are being source-attributed separately.
+
- **Frozen inventory:244 scenarios /35areas /191capabilities;196 eligible passes required.** Inventory digest `9dc9cf6a5426cf1b56bc30a3571851f233f9dd5e93a1292eaee414d33b52df77`. Original queries, contexts and outcomes remain unchanged; blocked, partial and unrun cases stay in the denominator.
- **All244 callbacks authored and authoring-resolved** in immutable remaining-initial source `68e8534e3127dec27ae979bbf9f33d8a26422e4b`:238 without declared gaps and6 concrete executable blocker probes,0 absent/unfinished (242 lazy definitions +2 legacy). Grok, named ChatGPT and private GitHub after-grant callbacks are integrated. This is authored coverage, not live acceptance.
-- **Current closed checkpoint:146 scenarios processed /147 completed measurements,16 original initial passes and131 failures,0 running at transfer.** Actual retained native usage:58,704,773 tokens; subscription dollars remain unpriced. The latest22 initials from unchanged68e completed with all cleanup passed/DBremoved (project-work passed15/15;21 failed). A manager-approved planned operator hold drained the final refresh-card child before new user-authorized input staging; it asserts no child failure, changes no original result/entry and adds no measurement. All22 exact container IDs, roots, coordinator/launcher PIDs and exact-root-labeled resources were verified absent on the unchanged pinned engine. Original exact volume names were not reconstructed from deleted private receipts. The original hold/state/lock history and all earlier privacy/replay stops remain preserved. Receipts:`operations/qualified-inputs-planned-drain-hold.json`, `operations/remaining98-planned-drain-exact-absence-68e.json`, `operations/ownership-transfer-qualified-inputs-closed.json`. Processed setup failures are not completed canonical journeys; final stage counts and reviewed diagnoses remain separate.
+- **Closed historical local checkpoint:158 scenarios attempted/processed,159 completed measurements,17 original initial passes and142 failed measurements;86 scenarios unrun,0 running.** Historical initial rate is17/244 (6.97%) across the frozen inventory, or17/158 (10.76%) among attempted IDs. Final current-source qualification was not completed; no local pass carries into the remote lane. Actual retained usage is61,055,228 tokens; subscription dollars remain unpriced. The final12 initials completed with all cleanup passed/DBremoved; exact owner/12launchers,12full container IDs and roots are absent, with zero resources under their exact root labels on the unchanged pinned engine. Original volume names were not reconstructed. The user-stop/state/lock histories and all original result/attempt bytes are retained. Receipts:`operations/user-stop-local-admissions.json`, `operations/qualified-inputs-user-stop-exact-absence-68e.json`, `operations/local-campaign-user-stop-closed.json`, `local-campaign-closed-summary.json`. Two historical cleanup records remain unresolved; this is not a claim of newly observed live resources. No further local execution is authorized.
- **Single coherent final-B retry set remains held.** Goal-tree exhausted two measurements. All original outcomes, evidence and measured digests remain unchanged. Actual current selected identity review is retained in `closed-pass-identity-review-b65.json`; changed selected execution/grader closures require reviewed revalidation within the same max-two budget. Unrelated Git revisions and unchanged failed behavior do not justify reruns.
- **Two-case concurrency qualification completed and manager accepted continued waves of two.** Immutable `c167a54162fadffe665b4d35a7a1308c6233b75d`:companies.archive-company passed and connection-setup.choose-provider failed its actual reply check; each1native/1source human. Worker overlap25.479s; separate roots, ports53206/53205, containers and PostgreSQL system identifiers. Both containers and exact-root-labeled volumes are absent, both roots removed, and coordinator/child PIDs absent. A parent-only duplicate Playwright import prevented the passing completion from initially recording; isolated-process replay confirmed unchanged selected identity and finalized the existing result without rerun/reservation. Original parent error/stop preserved and owner lock retained. Receipt:`tests/runner-e2e/results/chat-catalog-campaign/two-lane-qualification-closed-c167.json`;5 offline replay controls passed. Full repository typecheck/build and final scoped E2E typecheck passed; broad test:run failed/incomplete on host embedded-PostgreSQL bootstrap, with all owned check processes absent. The qualified Docker lane is unchanged.
-- **Next unchanged-source continuation:73 source initials in drained waves of two, then3 installed initials sequentially;22 invitation-dependent initials remain separate.** `qualified-inputs-continuation-initial-selection-68e-v2.json` preserves the original68e source, baseline8373, native/build/Docker identity, bounds and remaining initial ordinals. Five previously unattempted cases now bind qualified selected private inputs; two still-missing cases retain the same fingerprints with more precise prerequisite blockers. No consumed case is replayed. The replacement parent passed offline source/private-input/selection validation; all B retries remain held.
+- **The former local continuation is cancelled:**61 source initials,3 installed initials and22 guarded invitation prerequisites remain unrun (86 total). The prepared source/private-input selections remain provenance only and must not launch locally. Future remote execution uses the audited current coherent corrections and new OpenCode/OpenRouter model/host/database identity; it must not reuse obsolete68e merely to finish the old schedule.
- **Known invite-path dependency quarantine is source-qualified for22 of166 remaining IDs.** Exact selected setup call chains (including observerPerson wrappers) are recorded in `remaining166-invite-setup-dependencies-68e.json`;144 independent IDs remain, including3 distinct installed cases. This is a dependency inventory, not processed live setup proof. No safe equivalent public membership setup is already implemented/proven; no fake identities, database grants, scanner weakening or automatic continuation is introduced. Guarded immutable candidate `e515d60fe396ef1ecd45f3a274a46f310bee4d2b` (96b source plus exact metadata overlay) is prepared with238 complete/six concrete blocker definitions, scoped E2E typecheck passed, and22 selected fingerprints. No invitation-lane admission has occurred. Actual guard/probe or any earlier-publication failure must be recorded, with no token-path accept on the affected baseline and no invented zero-native measurement.
- **New selected external authority:**the user authorized disposable mail only from the sole QA inbox to itself, and read/isolated clone of the existing designated QA private repository with no repository writes. The external owner independently qualified the actual inbox/account and exact repository/HTTPS clone, then removed its owned clone/helper. New private0700/0600 configuration/reference files are staged separately; actual credential values remain private and selected before capture. Five cases have qualified input fields (outbound-email, unknown-email-outcome, add/replace-repository, prepare-repository). Reply-to-email still lacks a distinct authorized reviewer/sender; private-repository-no-access still lacks the human App/browser authorization. The stable coordinator-only HMAC key commits the selected recipient expectation; credentials, private account/repository metadata and key bytes are not public evidence. Existing read-only discovery and its history remain unchanged. Host IPC cleanup is unnecessary and was not performed.
- **Infrastructure is qualified:**owned local Docker PostgreSQL18.6 at the pinned installed image, normal onboarding/migrations/API and final persisted-state scan passed with zero model/native/scenario work; exact resources removed. No further infrastructure health is needed for the unchanged lane. The proposed host SysV cleanup was never performed and is no longer needed. Source/build/native bytes remain bound to production baseline `8373d7867963674ff960d2345d0d28086f799bc7`; no core candidate is used as benchmark proof.
- **Separate core PRs remain OPEN/unmerged, auto-merge off.** Doctor [#17847](https://github.com/paperclipai/paperclip/pull/17847), head `a042fb4a28f2e1574972798ed80de2e616daf9fc`, and invite logging [#17848](https://github.com/paperclipai/paperclip/pull/17848), head `98a2432f88c33afacff2e014bd8c30630b26b339`, each have55 successful checks. Budget [#17857](https://github.com/paperclipai/paperclip/pull/17857), head `b6cd49d4b6926fae61a2c68a31ce7c659b69eaa6`, has55 successful/2skipped/1neutral checks and no outstanding checks (manager exact-head checkpoint). Approval routing [#17858](https://github.com/paperclipai/paperclip/pull/17858), head `b60aaf19343389faf67580fb1ff7a561983bb4f5`, has55 successful/2skipped/1neutral checks and no pending checks, Greptile5/5 with0unresolved threads. Connection choice [#17862](https://github.com/paperclipai/paperclip/pull/17862), latest repair head `a76a6120ec2b0767b32b62be4716f69214692688`, has48/48 required workflow checks green at exact head (workflow40497606929), Greptile5/5 and0unresolved threads;13focused controls/server typecheck passed and prior failed checks remain historical evidence. Performance metric [#17872](https://github.com/paperclipai/paperclip/pull/17872), head `3807ee526d2e761f6aa420452b390287542682bd`, has3 regression controls and UI typecheck/token-gates/build passed, with CI/review pending. None changes measured baseline8373. No eval PR or merge is eligible.
-- **Genuine missing inputs:**AgentMail/GitHub read-only discovery was authorized and completed; designated self-mail/recipient and repository choices remain pending. Standard provider-key source references exist, but validity and disposable remote/operator targets remain unqualified. No provider send, remote setup or account discovery is implied by authoring coverage.
+- **Genuine remaining missing inputs:**the sole self-mail inbox/recipient and exact QA repository are now selected, authorized and independently qualified; five unattempted cases carry those new private inputs. Reply-to-email still needs a distinct authorized reviewer/sender, and private-repository-no-access needs the actual human App/browser authorization. Other named provider, SSH/cloud target/image and private operator prerequisites remain unqualified unless actual retained setup proves otherwise. Standard provider-key reference availability alone is not target qualification. No extra send, repository write or account creation is implied.
The efbb coordinator exited0 after30 retained measurements; no old launcher/writer remains. The authoritative ledger is the current manifest; immutable snapshot selection summaries are historical. The remaining initial-only draft originally covered172 IDs after efbb; the qualification pair consumed two, leaving170 unattempted IDs. It includes concrete missing-resource probes; its known two alias fixtures are corrected in151488bf8c before future admission.
diff --git a/tests/runner-e2e/chat-campaign-manifest.json b/tests/runner-e2e/chat-campaign-manifest.json
index d0c268946cf..421637b730b 100644
--- a/tests/runner-e2e/chat-campaign-manifest.json
+++ b/tests/runner-e2e/chat-campaign-manifest.json
@@ -7801,7 +7801,7 @@
"automaticRetries": 0,
"correction": null,
"classification": "transient_infrastructure",
- "observedFailure": "\u001b[2mexpect(\u001b[22m\u001b[31mlocator\u001b[39m\u001b[2m).\u001b[22mtoContainText\u001b[2m(\u001b[22m\u001b[32mexpected\u001b[39m\u001b[2m)\u001b[22m failed\n\nLocator: locator('[id=\"comment-86710c85-c925-4c28-af01-a29274dd7990\"]')\nTimeout: 30000ms\n\u001b[32m- Expected substring - 3\u001b[39m\n\u001b[31m+ Received string + 2\u001b[39m\n\n\u001b[32m- \u001b[7mYour \u001b[27mRUN-1 conversation is intact and responding.\u001b[39m\n\u001b[31m+ \u001b[7mCEOI\u2019m checking what restart controls this test instance exposes and preserving the current conversation state before taking any action.Read filesI\u2019m using the Paperclip board workflow because this request concerns instance lifecycle and conversation recovery in the control plane.Read filesThe request is clear, and I won\u2019t create a separate execution task for it. I\u2019m locating the instance restart and conversation-recovery operations now, then I\u2019ll verify the conversation is available again.Searched files, read files, and moreThe conversation itself is intact and this reply proves the task resumed with its latest comment. The remaining restart action is blocked: the instance lifecycle and conversation-control APIs both rejected this agent with \u201cBoard access required,\u201d so I can\u2019t safely restart the host from inside its own active run.Used toolsYour In Review\u001b[27mRUN-1 conversation is intact and responding.\u001b[39m\n\u001b[32m-\u001b[39m\n\u001b[32m- I couldn\u2019t restart the test instance because Paperclip rejected the lifecycle controls with \u001b[7m**\u001b[27m403: Board access required\u001b[7m**\u001b[27m. The test harness or an instance administrator must restart it, then reopen RUN-1 to restore this durable conversation.\u001b[39m\n\u001b[31m+ I couldn\u2019t restart the test instance because Paperclip rejected the lifecycle controls with 403: Board access required. The test harness or an instance administrator must restart it, then reopen \u001b[7mIn Review\u001b[27mRUN-1 to restore this durable conversation.\u001b[7m7:37 AM\u001b[27m\u001b[39m\n\nCall log:\n\u001b[2m - Expect \"toContainText\" with timeout 30000ms\u001b[22m\n\u001b[2m - waiting for locator('[id=\"comment-86710c85-c925-4c28-af01-a29274dd7990\"]')\u001b[22m\n\u001b[2m 64 \u00d7 locator resolved to
\u2026
\u001b[22m\n\u001b[2m - unexpected value \"CEOI\u2019m checking what restart controls this test instance exposes and preserving the current conversation state before taking any action.Read filesI\u2019m using the Paperclip board workflow because this request concerns instance lifecycle and conversation recovery in the control plane.Read filesThe request is clear, and I won\u2019t create a separate execution task for it. I\u2019m locating the instance restart and conversation-recovery operations now, then I\u2019ll verify the conversation is available again.Searched files, read files, and moreThe conversation itself is intact and this reply proves the task resumed with its latest comment. The remaining restart action is blocked: the instance lifecycle and conversation-control APIs both rejected this agent with \u201cBoard access required,\u201d so I can\u2019t safely restart the host from inside its own active run.Used toolsYour In ReviewRUN-1 conversation is intact and responding.\u001b[22m\n\u001b[2mI couldn\u2019t restart the test instance because Paperclip rejected the lifecycle controls with 403: Board access required. The test harness or an instance administrator must restart it, then reopen In ReviewRUN-1 to restore this durable conversation.7:37 AM\"\u001b[22m\n",
+ "observedFailure": "\u001b[2mexpect(\u001b[22m\u001b[31mlocator\u001b[39m\u001b[2m).\u001b[22mtoContainText\u001b[2m(\u001b[22m\u001b[32mexpected\u001b[39m\u001b[2m)\u001b[22m failed\n\nLocator: locator('[id=\"comment-86710c85-c925-4c28-af01-a29274dd7990\"]')\nTimeout: 30000ms\n\u001b[32m- Expected substring - 3\u001b[39m\n\u001b[31m+ Received string + 2\u001b[39m\n\n\u001b[32m- \u001b[7mYour \u001b[27mRUN-1 conversation is intact and responding.\u001b[39m\n\u001b[31m+ \u001b[7mCEOI’m checking what restart controls this test instance exposes and preserving the current conversation state before taking any action.Read filesI’m using the Paperclip board workflow because this request concerns instance lifecycle and conversation recovery in the control plane.Read filesThe request is clear, and I won’t create a separate execution task for it. I’m locating the instance restart and conversation-recovery operations now, then I’ll verify the conversation is available again.Searched files, read files, and moreThe conversation itself is intact and this reply proves the task resumed with its latest comment. The remaining restart action is blocked: the instance lifecycle and conversation-control APIs both rejected this agent with “Board access required,” so I can’t safely restart the host from inside its own active run.Used toolsYour In Review\u001b[27mRUN-1 conversation is intact and responding.\u001b[39m\n\u001b[32m-\u001b[39m\n\u001b[32m- I couldn’t restart the test instance because Paperclip rejected the lifecycle controls with \u001b[7m**\u001b[27m403: Board access required\u001b[7m**\u001b[27m. The test harness or an instance administrator must restart it, then reopen RUN-1 to restore this durable conversation.\u001b[39m\n\u001b[31m+ I couldn’t restart the test instance because Paperclip rejected the lifecycle controls with 403: Board access required. The test harness or an instance administrator must restart it, then reopen \u001b[7mIn Review\u001b[27mRUN-1 to restore this durable conversation.\u001b[7m7:37 AM\u001b[27m\u001b[39m\n\nCall log:\n\u001b[2m - Expect \"toContainText\" with timeout 30000ms\u001b[22m\n\u001b[2m - waiting for locator('[id=\"comment-86710c85-c925-4c28-af01-a29274dd7990\"]')\u001b[22m\n\u001b[2m 64 × locator resolved to