Current models
0Set the models you use now so later changes have a clear baseline.
Manage current modelsExpired device records are removed when this saved-records view is opened. A closed browser cannot perform scheduled deletion.
ACCOUNT & SUBMISSIONS
Manage your browser-local AI Identity, strategies, model candidates, evaluation progress, and important change signals in one place. Nothing is uploaded automatically.
CLOUD MEMBER CENTER
Keep explicitly selected records in your account for access on another device. Existing local content is not migrated automatically, and no model is called.
Checking this cloud account…
Selected cloud content is retained until you actively delete it. Saving, uploading, downloading and deleting each require their own explicit action. Local content is not synchronized automatically.
The platform’s backup retention and deletion propagation times have not been verified. Removing online content does not prove that every backup or copy has been erased immediately.
Sign-in identifies the uploader, not the original author or historical confirmer. Hashes check integrity, not identity or model quality. Cloud records do not grant execution, acceptance or confirmation authority; no model is called.
A minimal deletion marker retains the account identifier and content hash to prevent a late upload from restoring this record. The marker is not a copy of its content. This exact record cannot be uploaded again in this version.
Limits: 20 active documents, 16 MiB total content, 8000000 bytes per document, and 200 document identities per account lifetime, including deleted records. Content is not silently truncated, replaced or evicted.
Save a snapshot only after explicit confirmation. Unsaved drafts stay temporary; earlier saved snapshots remain in history. Nothing is uploaded and no model runs.
Read-only history — not a new acceptance confirmation
Shows explicitly saved content in this browser, not every draft, trial or activity.
Reading saved local content…
A missing local reference is not an incomparable result. Comparability is checked separately using the exact saved materials.
Temporary unsaved work is not listed and cannot be recovered from this overview.
Saved confirmations remain historical claims, not a current confirmation. Hashes establish integrity only.
This overview is read-only: it does not save, upload, run models or choose conclusions. Each action still needs its existing confirmation.
Task-pack history and retrial summaries use separate full-content JSON backups. Export each explicitly; one does not back up the other.
Hashes check file integrity, not truthful claims, provider or model identity, execution or result quality. Imported confirmation times and confirmed-by fields are unauthenticated claims. Import transfers no approval, identity or authority.
Saved content uses localStorage for this site, browser and profile only. It is not cloud storage or on-device model inference. Clearing browser data, storage limits or device loss can remove it. Export a backup explicitly when needed.
The existing workspace backup does not include these new full-content records. Use the separate task-pack history export to back them up.
File, record and browser-storage limits apply. Reaching a limit does not automatically delete older history. Export what you need; a failed save or import does not count as a backup.
Loading saved history…
Saved sources, quotations and confirmations are record claims, not authenticated facts. Hashes check integrity, not truth, identity or result quality. Viewing history does not reconfirm it.
Choose a full summary JSON export. Metadata, Markdown and task-pack history backups cannot be restored here. Validation and preview do not save anything. Nothing is uploaded and no model runs.
Restoration preserves the earlier records and items that are not directly comparable. It adds read-only history, not an editable task or execution permission. Local storage is not cloud sync, encryption or guaranteed durable storage.
The imported confirmation, name and time remain historical claims, not your new acceptance or authenticated identity. Hashes check integrity, not truth, authorship, rights or result quality. Restoration does not judge which result is better.
Limits: 8,000,000 UTF-8 bytes and 4,000,000 characters per file, and at most 20 saved summaries within the collection's content limits. Existing records are not automatically overwritten or deleted.
Load saved summaries
Supports a task-pack history backup or a full task-pack JSON export. Metadata-only files and standalone acceptance files cannot be restored; acceptance records need their matching task pack in a history backup.
Continue each validated Decision Brief from public evidence to representative testing and change monitoring.
Loading your workspace…
Saved models and compact Radar or evaluation summaries are preserved for continuity. They are not Decision Workspace v2 history or current evidence.
No compact legacy decision summary is stored.
Set the models you use now so later changes have a clear baseline.
Manage current modelsStart a comparison or real-task evaluation to build a shortlist.
Continue shortlistSave models you want to watch without marking them as current.
Open saved modelsOpen the radar once to establish a change baseline.
Open decision radarSubmission review records are managed separately from local content and explicitly uploaded cloud records. Anonymizing submission identity does not delete cloud records.
Loading your workspace…
Optional defaults and memory references. This profile never overrides a Decision Brief or uploads automatically.
+PERSONAL MEMORY · LOCAL FIRST
Start with visible defaults, then review and save how you work, what you already use, and the boundaries every future strategy should respect.
Coverage counts visible selections, including suggested starting defaults. It does not measure performance or infer preferences.
13 of 17 configuration signals set
If this Site enables free cloud accounts, full records are stored only after you select each record and explicitly consent. Existing local content is not migrated automatically.
OpenGPT cannot read your ChatGPT conversations, files, memory, API keys, or billing details.
For account-linked actions, OpenGPT receives a site-specific user ID plus the available name and email.
Export official saved model IDs, legacy Radar settings, and the validated Decision Workspace v2 index, including versioned Brief, Workload, and Task Pack metadata. Active Decision history imports only as new or append-only records. Legacy evaluation prompts, pasted answers, notes, reports, and saves labeled Unresolved identity are excluded.