Advanced context retrieval for power users.

Continuity Engine PRO is the advanced version of Continuity Engine for power users who need deeper local transcript retrieval, prompt-time context injection, AI-assisted multi-session merge, and optional customer-keyed AI features.

Yes. Continuity Engine PRO is part of the current product ladder. It is monthly only at launch.

PRO includes the full Basic continuity system: local archive, transcript search, highlights, deterministic capsule, summary layer, Library document, multi-session updates, token-density controls, backups, and archiving.

Use PRO when you want the Engine to help retrieve relevant local project context during active prompting, not just during new-chat updates. It is for heavy project users, developers, researchers, founders, builders, and anyone whose AI work has become a living organism with folders.

No. PRO can include optional customer-keyed API features, but the customer chooses whether to enable them. API use is customer-paid and customer-controlled.

Start the local Continuity Engine first, then open a supported chat page and click the Chat Health floater in the lower-right corner of the chat. When the extension detects that the Engine is running, the floater switches to the Continuity Engine icon and opens the CE sidebar. If the Engine is installed but not running, the sidebar will not open because the extension stays in Health Meter or Schrodinger's Chat mode.

AI-operated context search lets the user enter a prompt or task, then have the PRO workflow search the local transcript archive for relevant context before the prompt is sent or continued. The goal is to inject the right memory at the right time.

Search depth controls how broad or deep the context retrieval should be. Shallow searches are faster and lighter. Deeper searches may find more relevant history but can take longer and, if API-backed, cost more.

Prompt-time injection means relevant project context can be added to the user's active prompt, not only to a one-time new-chat update. This helps a current AI session act with more of the project's actual history in view.

No. The design goal is selective retrieval, not blindly dumping the entire archive into an external model. If API-backed features are enabled, the customer controls the key and should use search-depth controls to manage what is sent and what it may cost.

Yes. PRO is designed for projects that span multiple sessions. The value increases when the project history is too large for a human to manually remember or re-copy.

It means the customer provides their own AI API key for optional features. Betweenware does not bundle API usage into the subscription. Any API charges belong to the customer's API account.

An API key can allow more intelligent search, merge, summarization, and retrieval behavior than deterministic local logic alone. It is useful when project history is large and the user wants an AI agent to help find the right context.

Where supported, local GPU resources can assist merge and search operations so more work can happen locally. GPU usefulness depends on the user's hardware and the specific feature.

Yes. A powerful GPU can help with some advanced local workflows, but it is not the entry requirement for using the basic PRO workflow. Optional GPU-backed features should be treated as enhancements, not the doorway.

Use selectable search depth, avoid unnecessary deep searches, and disable API-backed features when not needed. PRO should not be treated as a bottomless token blender.

PRO does not give Betweenware control over ChatGPT, Claude, Gemini, or any other AI provider interface. If those companies change their browser pages, some CE browser I/O features may require an update.

No. PRO is more capable, but users still need to understand the workflow. The FAQ answers common questions; the manual remains the operational source for exact installation and usage steps.

Send the product tier, version numbers, browser, operating system, whether API features were enabled, the AI workspace involved, the exact step that failed, and screenshots where possible.

Product account mechanics depend on the checkout and licensing system active at the time. For billing or plan-change help, email support.