@bootstrapware/answers 0.2.4 → 0.2.5

package/AGENTS.md

@@ -143,6 +143,8 @@ const answers = createAnswersBackend({
 export const POST = (request: Request) => answers.handler(request);
 ```
 
+Custom Q&A changes at runtime with no refresh: `await answers.upsertQa({ id, question, answer })`, `answers.removeQa(id)`, or `answers.setQa([...])`. Only the changed entry is embedded. To keep your own vector store, pass `retrieve` or call `answerQuestion` with your retrieval; custom Q&A is still always considered.
+
 The full reference app is `examples/answers-nextjs` (markdown knowledge on disk, a JSON visitor log).
 
 Citations must be ids you retrieved. The widget links a URL only when that URL is on the returned citation. Other URLs stay text. HTML in the answer stays text.
@@ -227,8 +229,8 @@ Hosted indexes the main body of each page, split by headings; navigation, header
 1. Make sure the fact is stated in a page body, not only implied by a homepage card. A pricing page, a support page, and one page per product work well.
 2. Keep each figure next to what it prices (`Hosted is $19.99/month`). An answer whose price or number is not in a cited passage is rejected.
 3. After `add_answers_url_source`, wait for the index job, then use the dashboard retrieval check to see which passages a question finds. Ask through the widget with a test key to see the written answer.
-4. For facts you want stated exactly (prices, refund policy, support contact), add a custom Q&A in the dashboard. Custom Q&A ranks ahead of page text.
-5. Unanswered is the correct outcome when the pages do not have the fact. Add a URL or a custom Q&A. Do not send the question through MCP.
+4. For facts you want stated exactly (prices, refund policy, support contact), add a custom Q&A in the dashboard. Custom Q&A is always considered, and a matching entry is answered first. Hosted entries are indexed on save, usually within about a minute; the dashboard knowledge list shows when each is indexed. With BYO, an entry changed through `upsertQa`, `removeQa`, or `setQa` applies from the next question.
+5. When the pages are silent on a yes/no question (certification, compliance, encryption), the answer says the site does not state it and gives the support contact if a source has one. Add the support contact as a custom Q&A so that works. Unanswered with the fallback is the correct outcome when nothing covers the question. Add a URL or a custom Q&A. Do not send the question through MCP.
 
 ## Knowledge, grounding, and provider
 
@@ -236,7 +238,7 @@ Hosted knowledge kinds are website root, sitemap, single public URL, PDF, DOCX,
 
 Every Hosted source is visitor-readable. Do not upload private runbooks or secrets. Crawl is public `http` and `https` only. Private hosts are rejected. `robots.txt` is honored. A failed crawl leaves the last good index in place. A page that is only navigation chrome is rejected and does not become a citation.
 
-Retrieval combines keyword and embedding search over the active index (custom Q&A first when it matches). The model answers in 1 to 4 sentences in the visitor's language, citing passages it was given; citations are ids from the retrieved index, and the model does not invent URLs. A draft that cites an unretrieved passage, states a figure no cited passage contains, or writes code not quoted from a passage is retried once and then returns `unanswered` with the configured fallback. Off-topic questions and prompt-injection attempts are declined before retrieval. There is no model-memory fallback. Follow-ups use the last three turns of the session.
+Retrieval combines keyword and embedding search over the active index. Custom Q&A is always considered, and a matching entry is answered first. The model answers in 1 to 4 sentences in the visitor's language, citing passages it was given; citations are ids from the retrieved index, and the model does not invent URLs. A draft that cites an unretrieved passage, states a figure no cited passage contains, or writes code not quoted from a passage is retried once and then returns `unanswered` with the configured fallback. Off-topic questions and prompt-injection attempts are declined before retrieval. How-to questions get the documented steps and a citation to the docs page; the model does not write new code. There is no model-memory fallback. Earlier turns are used only for clear follow-ups ("what about Chat?"); a standalone question is answered fresh.
 
 Unanswered observations store normalized question text even when conversation history is off, so the dashboard can group them. MCP does not return that text.
 

package/CHANGELOG.md

@@ -1,5 +1,9 @@
 # Changelog
 
+## 0.2.5
+
+- Docs only, no code changes. README and AGENTS.md match the v2 behavior: custom Q&A is always considered and a matching entry is answered first (replaces "ranks ahead of page text"); when the sources are silent the answer says the site does not state it and gives the support contact if a source has one; `answerQuestion` over your own retrieval and runtime Q&A changes; how-to questions cite the docs page without writing new code; earlier turns are used only for clear follow-ups. Hosted entries are indexed on save, usually within about a minute; with BYO, an entry changed through `upsertQa`, `removeQa`, or `setQa` applies from the next question.
+
 ## 0.2.4
 
 - Session context: earlier turns help only with clear follow-ups. Every question is routed on its own first, exactly as in a new session, and the router flags `followup` (a pronoun with no subject, "what about Chat?", "and the price?"). Only follow-ups are rewritten with the conversation (carrying over just the referent), and only they show earlier turns to the answer step; if the follow-up reading finds nothing, the question is answered on its own. A self-contained question asked late in a long session gets the same retrieval and answer as in a new session.

package/README.md

@@ -59,7 +59,7 @@ export const POST = (request: Request) => answers.handler(request);
 
 The index is built on the first question and kept in memory. Call `answers.refresh()` after your content changes, or save `answers.index.toJSON()` and pass it back as `snapshot` to skip re-embedding on restart. Pass `retrieve` to use your own vector store instead.
 
-Answers come only from the indexed passages. A draft that cites something it did not retrieve, or states a price or number no cited passage contains, is rejected and the visitor gets the fallback. Custom Q&A ranks ahead of page text.
+Answers come only from the indexed passages. A draft that cites something it did not retrieve, or states a price or number no cited passage contains, is rejected and the visitor gets the fallback. Custom Q&A is always considered, and a matching entry is answered first. When the sources are silent on a yes/no question, the answer says the site does not state it (with the support contact if a source has one). How-to questions cite the docs page; no new code is written.
 
 Custom Q&A can change at runtime without a refresh: call `await answers.upsertQa({ id, question, answer })`, `answers.removeQa(id)`, or `answers.setQa([...])` from your save handler. Only the changed entry is embedded and it answers on the next question. Relevant custom Q&A are always offered to the model, even when many page passages outrank them, and that includes setups that pass their own `retrieve`.
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@bootstrapware/answers",
-  "version": "0.2.4",
+  "version": "0.2.5",
   "description": "Embeddable source-backed Q&A for your site: React widget plus a server engine (crawl, hybrid search, grounded OpenAI answers with citations) for BYO backends.",
   "license": "MIT",
   "private": false,