PR #31536 real runtime verification
head: 803e6d0c1ca8bb50d9ea6b2a4a4b422a4fae2cf2
main: e4d86abca9c0d0ddd8107ea557f2e0c879632cf3


main: actual codex.app-server completed MCP calls
preview_status: completed, 135ms
{"available":false,"visible":false,"tabId":null,"url":null,"title":null,"loading":false,"profiles":[{"id":"default","name":"Default"},{"id":"incognito","name":"Incognito","incognito":true}],"defaultProfileId":"default"}
preview_open: completed, 15701ms
{"available":true,"visible":false,"tabId":"tab_1_c13e2d20-4096-472a-a095-62fcf89f696b","url":"http://localhost:8693/","title":"T3 Code (Beta)","loading":false,"control":{"owner":"agent","ownedByCaller":true,"generation":0},"dialog":null,"viewportSetting":{"_tag":"fill"},"viewport":{"width":1280,"height":800},"fileChooser":null,"downloads":[],"tabs":[{"tabId":"tab_1_c13e2d20-4096-472a-a095-62fcf89f696b","url":"http://localhost:8693/","owner":"agent","ownedByCaller":true,"visible":false,"profileId":"default"}],"profiles":[{"id":"default","name":"Default"},{"id":"incognito","name":"Incognito","incognito":true}],"defaultProfileId":"default"}
t3_preview_close: completed, 21ms
{}

pr: actual codex.app-server completed MCP calls
preview_status: completed, 135ms
{"available":false,"visible":false,"tabId":null,"url":null,"title":null,"loading":false,"profiles":[{"id":"default","name":"Default"},{"id":"incognito","name":"Incognito","incognito":true}],"defaultProfileId":"default"}
preview_open: completed, 15701ms
{"available":true,"visible":false,"tabId":"tab_1_c13e2d20-4096-472a-a095-62fcf89f696b","url":"http://localhost:8693/","title":"T3 Code (Beta)","loading":false,"control":{"owner":"agent","ownedByCaller":true,"generation":0},"dialog":null,"viewportSetting":{"_tag":"fill"},"viewport":{"width":1280,"height":800},"fileChooser":null,"downloads":[],"tabs":[{"tabId":"tab_1_c13e2d20-4096-472a-a095-62fcf89f696b","url":"http://localhost:8693/","owner":"agent","ownedByCaller":true,"visible":false,"profileId":"default"}],"profiles":[{"id":"default","name":"Default"},{"id":"incognito","name":"Incognito","incognito":true}],"defaultProfileId":"default"}
t3_preview_close: completed, 21ms
{}
preview_status: completed, 47ms
{"available":false,"visible":false,"tabId":null,"url":null,"title":null,"loading":false,"profiles":[{"id":"default","name":"Default"},{"id":"incognito","name":"Incognito","incognito":true}],"defaultProfileId":"default"}
preview_open: completed, 5665ms
{"available":true,"visible":false,"tabId":"tab_1_b4732726-0e97-4e8a-80d7-899fdc8233b3","url":"http://localhost:5733/","title":"T3 Code (Beta)","loading":false,"control":{"owner":"agent","ownedByCaller":true,"generation":0},"dialog":null,"viewportSetting":{"_tag":"fill"},"viewport":{"width":1280,"height":800},"fileChooser":null,"downloads":[],"tabs":[{"tabId":"tab_1_b4732726-0e97-4e8a-80d7-899fdc8233b3","url":"http://localhost:5733/","owner":"agent","ownedByCaller":true,"visible":false,"profileId":"default"}],"profiles":[{"id":"default","name":"Default"},{"id":"incognito","name":"Incognito","incognito":true}],"defaultProfileId":"default"}
t3_preview_close: completed, 28ms
{}
preview_status: completed, 10ms
{"available":false,"visible":false,"tabId":null,"url":null,"title":null,"loading":false,"profiles":[{"id":"default","name":"Default"},{"id":"incognito","name":"Incognito","incognito":true}],"defaultProfileId":"default"}
preview_status: failed, 15ms
MCP credential does not grant the preview capability: browser preview tools are off for this thread. Do not retry them. To check a page, use a headless browser from the shell, such as Playwright, or curl. The user can turn on "Agent browser access" in Settings; it applies when the agent session next starts.

Grader commands, run on Blacksmith:
vp test run evals/mcp-tool-use/run.test.ts
PR: 510 passed.
Main runner against PR regression tests: 3 script-source checks failed; combined maxCalls alternatives failed; browser-access-off mock failed. PR runner restored and all 510 passed again.

Real ordinary Codex turn on PR:
git rev-parse HEAD
803e6d0c1ca8bb50d9ea6b2a4a4b422a4fae2cf2
git status --short
(no output)

No product files changed. The PR only adds eval cases and runner support; no Claude routing fix ships here.