# Jev deployment and live mirror controller feedback (2026-10-03) This is a live deployment record, not a completed speedrun claim. Run `mirror_20261003_142030` started at `2026-10-03T14:20:30.186+08:00`. The timer includes English UI compatibility work and all interruptions. ## Deployment - Updated main from `97eca20` to `95ae060`, preserving local AALC files. - Restarted `LcMCPServer` and validated a fresh MCP session, screenshot/OCR, and idle AALC task state. The desktop's old exposed transport stayed closed; fresh subprocess MCP sessions work without another desktop restart. - The runtime launcher now imports the repository controller and matching modules instead of retaining an older independent script copy. - Configured the actual `jev-1.13-free` choice endpoint at `https://opencode.ai/zen/v1/systemone`. Credentials stay in private runtime configuration and are excluded from this repository and feedback. - An isolated identical route probe took 5468 ms with curl and 1859 ms with urllib. Switched the live configuration to urllib. Actual subsequent route calls included 1313 and 1484 ms; there are also slow requests and timeouts. The isolated probe is not counted as a live game decision. ## Live findings and repairs 1. The restarted Steam game uses English labels. Added controller-side label compatibility for entry, gift selection, event choices, judgement and confirmation boundaries. Preserve the original OCR text for diagnostics. 2. `Receive` incorrectly matched the old substring check for `REC`, causing unnecessary event advancement. Match the recording label exactly. 3. English OCR merges adjacent `Very High` labels. Derive separate explicit rate spans; normalization is idempotent and does not invent a high rate where OCR only says low or normal. 4. English event outcome text was offered as clickable choices. Exclude `check`, `gain` and E.G.O Gift description lines from event buttons. 5. Repeated generic confirmation requests added API latency. Execute a sole validated confirmation directly. Route and branch choices still use Jev. 6. Route and sinner API requests now contain current relevant observations, rather than unrelated entry, fusion and party setup requirements. 7. Sinner candidates are constrained to the explicitly highest visible rate; explicit OCR rates precede ambiguous colour-only candidates. The log saves the provider's scored candidate scope. Zeros outside that scope represent constraints, not extra probabilities returned by the API. 8. Initial candidate pruning caused a probability-set validation error. Fixed the adaptation between scoped provider probabilities and router validation, and added an integration regression test. 9. An enabled Commence button initially competed with sinner selection, causing repeated selection. An enabled judgement button now takes priority. 10. Enter sometimes fails on event judgement. Added the actual observed button as its mouse fallback. The original fallback list omitted judgement. 11. A changing event video clock reset repeat detection. Ignore both colon and decimal clock formats when detecting unsuccessful confirmations. 12. After combat, OCR read the map heading as `Exploring Floor` or `Exploring FloorT` without a valid digit. The old normalizer failed to mark this as a map, so P/Enter batches continued on the map for several minutes. Recognize the map heading independently of its floor number and retain the previously known floor. Added a regression against continuing battle. 13. Shop OCR splits Enhance/Gifts and reads the skill panel as SkillsOwned. Normalize those labels and truncated Replace Outis's Skil. 14. The keyword panel has a second Refresh button. Stock refresh now uses the leftmost validated Refresh, rather than overwriting it with keyword refresh. 15. English skill payment uses Confirm instead of Purchase. Only offer payment after verifying the exact requested skill pair, with Enter and a mouse fallback. 16. Full-screen OCR omitted the orange Purchased badge despite successful payment. This caused attempts to reopen the same skill slot. Added a small observed label template and a regression for both a bought and an empty slot. 17. A sole validated skill option or stock refresh executes without an extra provider call. Store exit uses Enter, then the observed-button fallback. 18. Added English server-error/Retry labels and Cost/resource reward anchors. A server-error modal excludes the reward buttons behind it. 19. Added `show_game_background`, using SW_SHOWNOACTIVATE. Live recovery returned visible=true and foreground_changed=false after a hidden game restart. 20. Code inspection showed AALC BackgroundInput.set_mouse_pos moves the actual system cursor. MCP background clicks and drags now send client mouse messages directly. Unity's title modal did not respond to messages alone; the deployed `window_move` mode aligns an invisible game window with the existing cursor, waits for input processing and restores position without activation. Closing the error modal and clicking TOUCH TO START both responded in the real game. Battle and title-drag completion with this new mode remain unverified. ## Progress verified so far - Entry blessings: base 1, 3 and 10; no enhanced blessings. - Starter gifts: Barbed Lasso and Fluorescent Lamp. - Actual Battery and permitted Thunderbranch obtained. - Floor 1 pack: Nest, Workshop, and Technology, with Talisman Bundle in the displayed clear rewards. One pack refresh was used before finding it. - Kept the user's seven-character preset and used Enter to enter combat. - First battle completed, then two event nodes completed. At 15:27 the controller entered another battle using its P/Enter macro. - Floor 1 shop: actual Thunderbranch++ verified after payments of 75 and 150 (balance 590 -> 515 -> 365). Actual Outis 2->3 verified after paying 75 (balance 365 -> 290 and visible Purchased badge). - No final reward or dungeon completion has been verified in this record. The preceding interruptions do not qualify as a successful fast run. The controller sends background game inputs. There is one game executor; root interventions occur only after it hands off or stops. Battle batches use P/Enter, and deterministic confirmations and event advancement avoid per-key model decisions. Each action and timing remains in the private runtime JSONL log; no provider key is written into it. ## Connection failure and recovery boundary - At the next Floor 1 post-combat reward, repeated CONNECTING and server-error retries prevented further progress. The controller handed off at 16:13:42. - Restarted the client without resetting `mirror_20261003_142030`, its timer or the verified inventory flags. Saved the pre-restart run/context privately. - The title login still failed after restart and a further actual button retry. Player.log reports: `Curl error 35: Cert verify failed` and `UnityTls error code: 7`. This is an observed failure, not a proven root cause. - The game's proxy log identifies www.limbuscompanyapi.com. Independent HTTPS probes using the existing local proxy and direct connection both timed out after 10 seconds. No game API response or restored dungeon was verified. - At 16:48 the game was still showing its login server-error modal. The mirror controller remains stopped; no new dungeon or reward claim was fabricated. The user was asked whether they can restore a successful game login. - Validation: 171 unit/regression tests passed in 11.297 seconds. Live evidence covers title-modal input and no-activation window recovery, not a completed run. - Credentials, proxy configuration, full screenshots with account identifiers, and untracked AALC changes are excluded from this upload. - `mirror_20261003_142030-actions.jsonl` contains the real current-run action choices, engine attribution, probabilities and timings. Raw OCR, screenshots, private configuration and arbitrary action arguments are omitted. Test-run records and other runs are excluded; controller step numbers can restart after a resume and are not a unique action identifier.