D DialogueBaseLocal-first lab

GROUNDED ASSISTANTS

Inspect the knowledge,
not just the answer.

Build a cited support assistant from text you control. Every answer shows the exact source chunks it used.

Add a source

TXT, Markdown, CSV, JSON and HTML are parsed locally. No upload occurs.

PLAYGROUND

Ask, inspect, correct

Optional BYOK model pass

This release prepares requests but does not send provider calls or spend money.

Ask a question to see a deterministic local retrieval trace.

QUALITY REVIEW

Turn conversations into test cases

Add regression case

Evaluation suite

Conversation review

ANALYTICS

See the gaps behind the score

Question themes

Source coverage

Failure queue

Safety and handoff

Content gaps

Unanswered or unresolved visitor questions, clustered by shared topic. Each gap is a source you could write.

SHARE

Put the assistant where the questions are

Public chat

Public chat answers only from your sources. Visitors see citations when enabled; nothing leaves this browser in the local build.

Link and embed

Captured leads

Names and emails visitors typed into the public chat on this browser. Export and follow up yourself - nothing is emailed automatically.

TRUST

The boundary is visible

Local retrieval

This release uses lexical scoring in your browser, not an external model. It never claims semantic AI accuracy.

Injection warnings

Source text is scanned for instruction-like phrases and flagged for review, not silently obeyed.

Portable data

Export all sources, chunks, conversations and evaluations. Delete them in one action.