@colixsystems/widget-sdk 0.244.0 → 0.245.0

package/README.md

@@ -55,7 +55,7 @@ The data layer lives in **four separate domain-client packages**, each instantia
 | **DATASTORE** | `useDatastoreSchema(tableId)` | `{ schema, loading, error, refetch }` | `schema(tableId)` — `datastore.read:<table>` |
 | **DATASTORE** | `useBoundColumns(tableId, shape, props)` | `{ columns, resolved, missing, loading, error }` | `schema(tableId)` (built on `useDatastoreSchema`) — `datastore.read:<table>`. Resolves author-bound column NAMES from `props` by exact name → case-insensitive name → first unclaimed column matching `shape[key].dataType`, so a column an author renamed after install still resolves instead of `record[props.titleField]` reading `undefined`. `columns` holds the resolved NAME (`record[columns.titleField]`); `resolved` holds the full `Column`; `missing` lists non-`optional` keys that never resolved. Falsy `tableId` collapses to `{ columns: {}, resolved: {}, missing: Object.keys(shape), loading: false, error: null }`. |
 | **DATASTORE** | `useInterpretDraft(tableId)` | `{ interpret, interpreting, error, result, available }` | `interpret(tableId, body)` — `datastore.read:<table>`. Turns ONE sentence a user typed ("walk at 11 am tomorrow") into DRAFT column values so a form can prefill itself. IMPERATIVE: call `interpret(text, { fields?, timeZone? })` from an event handler, never on mount. It DRAFTS and writes nothing — show the values for review, then submit through `useDatastoreMutation().create`. Resolves to `{ values, unresolved }`; `values` is keyed by column NAME (the shape `create()` takes) and `unresolved` names the fields the sentence did not state. Only text / number / boolean / date / datetime / array columns are drafted — `FILE`, `RELATION`, `USER` and `USER_GROUP` carry ids and are never guessed. Fails closed to an empty draft. **Every call spends the workspace's AI credits** and is rate-limited per actor, so call it once per user action (never on mount or in a render loop); once the workspace runs out the call is refused with a generic 429 — an app user is deliberately **not** told the workspace's billing state, since they have never heard of an AI credit and cannot buy one. Never surface a raw error to the person filling the form: say drafting is unavailable and keep every field editable by hand. `available` is false where the host brokers no interpreter. |
-| **AGENTS** | `useAgent(agentRef)` | `{ send, cancel, reset, messages, sending, resuming, error, available }` | `ctx.agents` — `ai.invoke:*`. Hold a conversation with one of the WORKSPACE's own AI agents (an author-written name + system prompt). Sending is IMPERATIVE: call `send(text)` from an event handler, never on mount — the first send opens the conversation, so a widget nobody types into costs nothing. On mount the hook DOES reopen the caller's most recent thread with this agent, so a refresh or a navigation keeps the conversation instead of starting a blank one; that resume only READS and spends no credit. `resuming` is true while it runs — render a loading state, not an empty thread. `messages` is the running transcript, oldest first. You **cannot** set or read the system prompt, replay an edited transcript, or pick a model: the server assembles every turn from the stored prompt plus the stored history, and `content` is the only thing that reaches the wire. `reset()` drops the thread (a "new chat" affordance), and the new thread is the one a later mount resumes. **Every send spends the workspace's AI credits** and is rate-limited per caller AND per workspace. Rejects with `AgentError`; branch on `.code` — `AI_QUOTA_EXCEEDED` (429) is the workspace's credit ceiling and will NOT clear on a retry, unlike `RATE_LIMITED`, so never offer "try again" for it, and `AGENT_DISABLED` (409) means the author turned it off, so hide the affordance. Never surface a raw error or the workspace's billing state to an app user. `available` is false when no agent is bound or the host brokers none — render a visible unbound state, never a blank box. An assistant message that used a tool also carries **`activity`** — what the agent DID with the user's data: the table it read, or the table and columns it CHANGED (never the values, and never a row the permissions refused). **Render a write.** A reply saying "done!" is not evidence a row was written, and if the agent got it wrong the user has no other way to notice. A talk-only turn omits the key entirely. |
+| **AGENTS** | `useAgent(agentRef)` | `{ send, cancel, reset, messages, sending, resuming, error, available }` | `ctx.agents` — `ai.invoke:*`. Hold a conversation with one of the WORKSPACE's own AI agents (an author-written name + system prompt). Sending is IMPERATIVE: call `send(text)` from an event handler, never on mount — the first send opens the conversation, so a widget nobody types into costs nothing. On mount the hook DOES reopen the caller's most recent thread with this agent, so a refresh or a navigation keeps the conversation instead of starting a blank one; that resume only READS and spends no credit. `resuming` is true while it runs — render a loading state, not an empty thread. `messages` is the running transcript, oldest first. You **cannot** set or read the system prompt, replay an edited transcript, or pick a model: the server assembles every turn from the stored prompt plus the stored history, and `content` is the only thing that reaches the wire. `reset()` drops the thread (a "new chat" affordance), and the new thread is the one a later mount resumes. **Every send spends the workspace's AI credits** and is rate-limited per caller AND per workspace. Rejects with `AgentError`; branch on `.code` — `AI_QUOTA_EXCEEDED` (429) is the workspace's credit ceiling and will NOT clear on a retry, unlike `RATE_LIMITED`, so never offer "try again" for it, and `AGENT_DISABLED` (409) means the author turned it off, so hide the affordance. Never surface a raw error or the workspace's billing state to an app user. `available` is false when no agent is bound or the host brokers none — render a visible unbound state, never a blank box. An assistant message that used a tool also carries **`activity`** — what the agent DID with the user's data: the table it read (with the `search` text and `filters` it read with), or the table and columns it CHANGED (never the values, and never a row the permissions refused). **Render a write.** A reply saying "done!" is not evidence a row was written, and if the agent got it wrong the user has no other way to notice. A talk-only turn omits the key entirely. |
 | **DATASTORE** | `useDatastoreMutation(table)` | `{ create, update, delete }` | `records(table).{ create, update (PATCH), delete }` — `datastore.write:*`. **To CLEAR a cell, pass it empty** — `update(id, { when: null })` (also `""` or `[]`) empties that cell and it reads back as `null` (`[]` on an array column). There is no separate clear call, so never write a sentinel such as `0` or a placeholder date to mean "no value". `false` and `0` are values, not clears, and a REQUIRED column cannot be cleared. |
 | **DATASTORE** | `useDatastoreSubscription(table, handlers, options)` (optional: options) | `{ status }` — `"connecting" \| "live" \| "reconnecting" \| "fallback"` | `records(table).subscribe` — `datastore.read:<table>`. Live `onCreated` / `onUpdated` / `onDeleted` off the REQ-RT-07 socket; never throws, resolving to `{ status: "fallback" }` so the widget polls instead. A whole-table subscribe is gated on read-EVERY-row, because one envelope reaches every subscriber of the table — so for a table governed by per-record grants pass `options.scope`: `{ kind: "record", record_id }` for one row, or `{ kind: "parent", relation_column, record_id }` for the rows whose RELATION column points at that parent (the column must carry `inheritAcl`, else the subscribe reports `"fallback"`). For a live list of the signed-in user's own rows pass `{ kind: "granted" }`: every row they read through a grant to them or one of their groups, plus `onAccessGained(record)` / `onAccessLost({ id })` as rows enter and leave that set. Re-subscribes on the scope's VALUES, so a fresh object literal each render is fine. |
 | **DATASTORE** | `useTyping(tableId, recordId)` | `{ typingUserIds, setTyping, status }` — status `"connecting" \| "live" \| "unavailable"` | `records(table).typing` — `datastore.read:<table>`. Who else is typing in one record, over the same socket as `useDatastoreSubscription`. Call `setTyping(true)` on every keystroke (throttled to one signal per 3 s; `false` is sent after 5 s idle) and `setTyping(false)` on send, clear or blur. `typingUserIds` lists the OTHER users' ids; each drops out 6 s after their last signal. Needs a signed-in user who can read the record. Never throws: an anonymous visitor, a refused subscription or a host whose client lacks `typing` reads `{ typingUserIds: [], status: "unavailable" }`. Re-subscribes when either id changes. |
@@ -84,7 +84,11 @@ See the design reference for the full architecture: [`docs/architecture/widget-m
 
 ## Status
 
-`v0.244.0` — pre-publish. The package surface (types, function names, export paths) is the v1 contract; runtime behaviour for some hooks is stubbed (each hook documents what's wired and what isn't). It is **not yet published to npm**.
+`v0.245.0` — pre-publish. The package surface (types, function names, export paths) is the v1 contract; runtime behaviour for some hooks is stubbed (each hook documents what's wired and what isn't). It is **not yet published to npm**.
+
+### What's new in 0.245.0 (contract 1.199.0)
+
+**An agent's activity says what it searched for.** A read entry in an assistant `AgentMessage`'s `activity` may now carry `search` (the text the agent searched every text column for) and `filters` (`{ column, operator, value }` conditions it read with). These are the agent's own arguments, never the rows that came back. When an agent misses an answer, this shows whether it searched for the wrong words or in the wrong column. Additive: a read with neither omits both keys, so existing renderers are unchanged.
 
 ### What's new in 0.244.0 (contract 1.198.0)
 

package/dist/contract.cjs

@@ -2064,8 +2064,8 @@ const HOOKS = [
       "`messages` is the running transcript, oldest first, each " +
       "{ id, role, content, created_at } with role 'user' | 'assistant'. " +
       "An assistant message that used a tool also carries `activity`: what " +
-      "the agent DID — the table it read, or the table and columns it " +
-      "CHANGED. Render a write: a reply saying 'done!' is not evidence a " +
+      "the agent DID — the table it read and the search text and filters it " +
+      "read with, or the table and columns it CHANGED. Render a write: a reply saying 'done!' is not evidence a " +
       "row was written, and the user has no other way to notice a mistake. " +
       "A talk-only turn omits the key entirely. " +
       "You CANNOT set the system prompt, replay an edited transcript, " +
@@ -6159,7 +6159,9 @@ const CONTRACT = deepFreeze({
   //    useFileSignatures is described as the caller's own signatures.
   // 1.198.0: additive (sc-12088) — `useNavigation().scrollIntoView(ref)`
   //    scrolls the page to an element inside the calling widget on both hosts.
-  version: "1.198.0",
+  // 1.199.0: additive (sc-11952) — an `AgentActivity` read entry may carry
+  //    `search` and `filters`: what the agent searched for, never its rows.
+  version: "1.199.0",
   sharedTranslationKeys: SHARED_TRANSLATION_KEYS,
   hooks: HOOKS,
   primitives: PRIMITIVES,

package/dist/contract.js

@@ -2064,8 +2064,8 @@ const HOOKS = [
       "`messages` is the running transcript, oldest first, each " +
       "{ id, role, content, created_at } with role 'user' | 'assistant'. " +
       "An assistant message that used a tool also carries `activity`: what " +
-      "the agent DID — the table it read, or the table and columns it " +
-      "CHANGED. Render a write: a reply saying 'done!' is not evidence a " +
+      "the agent DID — the table it read and the search text and filters it " +
+      "read with, or the table and columns it CHANGED. Render a write: a reply saying 'done!' is not evidence a " +
       "row was written, and the user has no other way to notice a mistake. " +
       "A talk-only turn omits the key entirely. " +
       "You CANNOT set the system prompt, replay an edited transcript, " +
@@ -6155,7 +6155,9 @@ const CONTRACT = deepFreeze({
   //    useFileSignatures is described as the caller's own signatures.
   // 1.198.0: additive (sc-12088) — `useNavigation().scrollIntoView(ref)`
   //    scrolls the page to an element inside the calling widget on both hosts.
-  version: "1.198.0",
+  // 1.199.0: additive (sc-11952) — an `AgentActivity` read entry may carry
+  //    `search` and `filters`: what the agent searched for, never its rows.
+  version: "1.199.0",
   sharedTranslationKeys: SHARED_TRANSLATION_KEYS,
   hooks: HOOKS,
   primitives: PRIMITIVES,

package/dist/index.d.ts

@@ -1790,6 +1790,17 @@ export interface AgentActivity {
   table?: string;
   record_id?: string;
   fields?: string[];
+  /** A read's own search text: what the agent looked FOR, never what it found. */
+  search?: string;
+  /** A read's own filter conditions, as the agent asked for them. */
+  filters?: AgentActivityFilter[];
+}
+
+/** One condition a read filtered by; `operator` is eq, neq, gt, gte, lt, lte or contains. */
+export interface AgentActivityFilter {
+  column: string;
+  operator: string;
+  value: string;
 }
 
 export interface AgentApi {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@colixsystems/widget-sdk",
-  "version": "0.244.0",
+  "version": "0.245.0",
   "description": "Common widget interface for AppStudio. Implements WidgetManifest, WidgetContext, property schema, and helper hooks.",
   "homepage": "https://github.com/Colix-AB/AppStudio",
   "type": "module",

package/src/contract.js

@@ -2064,8 +2064,8 @@ const HOOKS = [
       "`messages` is the running transcript, oldest first, each " +
       "{ id, role, content, created_at } with role 'user' | 'assistant'. " +
       "An assistant message that used a tool also carries `activity`: what " +
-      "the agent DID — the table it read, or the table and columns it " +
-      "CHANGED. Render a write: a reply saying 'done!' is not evidence a " +
+      "the agent DID — the table it read and the search text and filters it " +
+      "read with, or the table and columns it CHANGED. Render a write: a reply saying 'done!' is not evidence a " +
       "row was written, and the user has no other way to notice a mistake. " +
       "A talk-only turn omits the key entirely. " +
       "You CANNOT set the system prompt, replay an edited transcript, " +
@@ -6155,7 +6155,9 @@ const CONTRACT = deepFreeze({
   //    useFileSignatures is described as the caller's own signatures.
   // 1.198.0: additive (sc-12088) — `useNavigation().scrollIntoView(ref)`
   //    scrolls the page to an element inside the calling widget on both hosts.
-  version: "1.198.0",
+  // 1.199.0: additive (sc-11952) — an `AgentActivity` read entry may carry
+  //    `search` and `filters`: what the agent searched for, never its rows.
+  version: "1.199.0",
   sharedTranslationKeys: SHARED_TRANSLATION_KEYS,
   hooks: HOOKS,
   primitives: PRIMITIVES,