chore(web,tools,scripts): drop ticket references from comments (CL-8317)

agents/assistant/src/index.ts

@@ -9,13 +9,13 @@
 // platform's deploy machinery; the execution host materializes it at
 // runtime from the deploy alone.
 //
-// Tool-package pins (CL-5999, CL-5852): `@intx/agent`'s `defineAgent`
+// Tool-package pins: `@intx/agent`'s `defineAgent`
 // still does not accept a `toolPackagePins` field on its authoring-time
 // config — it is vendored, read-only source for this change — so the
 // agent below is built directly against `AgentDefinition`'s own type,
 // which already carries the field, matching
 // `workflows/collateral-generation`'s precedent. `@corbits/memory`
-// (CL-8186) is pinned so this deployment can search, add, and list the
+// is pinned so this deployment can search, add, and list the
 // tenant's firm memory (`memory_search`/`memory_add`/`memory_list`) — a
 // published dependency, not a workbench-built tool package, resolved
 // through its own `interchange.tools` surface the same way any other

agents/assistant/src/system-prompt.ts

@@ -1,7 +1,7 @@
 // Myra's pure prompt/pin data — no runtime import of `@intx/workflow` or
 // `@intx/agent` (only a type-only import below), so this module is safe
-// to bundle into a browser build (CL-8154's `apps/web` needs exactly
-// this: the content, not `buildAssistantWorkflow`'s `defineWorkflow`
+// to bundle into a browser build (`apps/web` needs exactly this: the
+// content, not `buildAssistantWorkflow`'s `defineWorkflow`
 // call, which pulls in `@intx/agent`'s Node-bound runtime — file
 // locking, in particular — that a browser bundle cannot resolve).
 // `index.ts`'s `buildAssistantWorkflow` imports these same constants;
@@ -18,7 +18,7 @@ export const ASSISTANT_STEP_ID = "assistant";
  * these: `@corbits/memory` is a normal published dependency, not a
  * workspace tool package the pin/registry resolver can name, so its
  * tool factories are attached directly on `toolFactories` in
- * `./index.ts` instead (CL-8186). `@corbits/capability-tools` lets Myra
+ * `./index.ts` instead. `@corbits/capability-tools` lets Myra
  * self-service a missing tool, skill, or model; the manager-tools
  * bundles give Myra real workbench-management capability — a specialist
  * agent she can create (each gets their own chat), connection
@@ -63,7 +63,7 @@ const ASSISTANT_WELCOME_CLAUSE =
   "allowed to.";
 
 /**
- * TRIAGE (CL-6350): how Myra decides, on every message, whether to
+ * TRIAGE: how Myra decides, on every message, whether to
  * answer directly or delegate — what to do when a job needs a
  * connection this workbench doesn't have yet, and the reverse case: she
  * answers by default and only steps back when a message @-mentions a
@@ -95,14 +95,14 @@ const ASSISTANT_TRIAGE_CLAUSE =
   "unless you are @-mentioned too or the sender asks you directly.";
 
 /**
- * TEAMMATE (CL-7469 folds in COORDINATOR): how Myra offers help without
+ * TEAMMATE (folds in COORDINATOR): how Myra offers help without
  * pushing it — folds in the skills-capture nudge rather than a separate
  * always-on clause. Myra is the coordinator of this bench's agents, not
  * just its resident assistant: she never builds a teammate by
  * hand-editing files or answering in its place, and never stands up a
  * second instance of an agent that already exists.
  *
- * The build-arc sentence (CL-5879) is the load-bearing one: it never
+ * The build-arc sentence is the load-bearing one: it never
  * waits for someone to name the mechanism ("make an agent", "set up a
  * routine"). An *outcome* — a sales motion, a content pipeline, a repo
  * to keep up — is enough on its own for Myra to work out the
@@ -145,7 +145,7 @@ const ASSISTANT_TEAMMATE_CLAUSE =
   "built and what it can now do once it's live.";
 
 /**
- * DISCOVERY (CL-6179): the specific interview-then-propose procedure a
+ * DISCOVERY: the specific interview-then-propose procedure a
  * stated outcome triggers, on top of the general "work out the team
  * yourself" doctrine in `ASSISTANT_TRIAGE_CLAUSE` above — short and
  * concrete rather than open-ended, so discovery never turns into a

agents/assistant/test/manager-tools-scenario.test.ts

@@ -1,5 +1,4 @@
-// A scripted playback of the owner's canonical GTM scenario (CL-5879
-// manager-tools follow-up), exercised directly against the manager-tools
+// A scripted playback of the owner's canonical GTM scenario, exercised directly against the manager-tools
 // bundles' `run()` with a single fake `fetch` standing in for every
 // workflow-run-authenticated route those bundles call. This proves the
 // CONTRACT across packages — a tool call's request shape matches what
@@ -132,9 +131,9 @@ function createFakeHub() {
     }
 
     if (url.pathname === "/api/workflow-chat/participants/messages" && method === "POST" && body) {
-      // The connect-service card post `request_connection` (CL-6393)
-      // makes into the caller's own room. Recorded so the scenario can
-      // assert the card itself, not just the tool's advice text.
+      // The connect-service card post `request_connection` makes into
+      // the caller's own room. Recorded so the scenario can assert the
+      // card itself, not just the tool's advice text.
       const parts = body["parts"] as {
         kind: string;
         block: { type: string; data: Record<string, unknown> };
@@ -244,7 +243,7 @@ async function runScenario(hub: ReturnType<typeof createFakeHub>): Promise<void>
   // A connector name that is neither a fixed `CONNECTOR_REGISTRY`
   // entry, a curated MCP preset, nor a connected MCP server makes
   // `request_connection` tell the agent to keep helping with what it
-  // can do (CL-6393) — never "go add a server and report back".
+  // can do — never "go add a server and report back".
   const unknown = await connectionsBundle.run(
     call("req_unknown", REQUEST_CONNECTION_TOOL, { connector: "acmecrm" }),
     new AbortController().signal,
@@ -255,7 +254,7 @@ async function runScenario(hub: ReturnType<typeof createFakeHub>): Promise<void>
   expect(hub.postedCards).toHaveLength(0);
 
   // She puts a connect card in the room for each of the three (real)
-  // services (CL-6393) — the card is what the human clicks; the tool
+  // services — the card is what the human clicks; the tool
   // result only tells Myra to keep helping in the meantime.
   for (const connector of ["exa", "granola", "linear"]) {
     const result = await connectionsBundle.run(

apps/web/README.md

@@ -57,7 +57,7 @@ Running routines in the contextual column are sourced today from
 | `/c`         | Workbench deep-link surface. On wide layouts the conversation opens in the right canvas; on compact layouts it fills the main pane. Legacy `/chat` links still resolve here. |
 | `/workflows` | Workflow runs executing across your benches.                                                                                                                                 |
 | `/library`   | The artifact gallery. See "Library" below.                                                                                                                                   |
-| `/settings`  | Account, bench membership, and (CL-5990) Agents/Skills. See "Agents" and "Skills" below.                                                                                     |
+| `/settings`  | Account, bench membership, and Agents/Skills. See "Agents" and "Skills" below.                                                                                               |
 
 Agents and Skills are Settings sections, not rail destinations — old
 `/agents[/:id]` and `/skills[/:id]` links still resolve, redirecting to

apps/web/src/agent-chat-launch.ts

@@ -2,9 +2,9 @@
 // fresh chat with it" — the same `POST /workbenches` call this app's every
 // create path uses. `CreateAgentPanel`'s Settings → Agents entry point
 // calls this on success so an explicitly-defined new agent never ends
-// nowhere, and `instant-agent-create.ts` — THE one creation verb
-// (CL-6138) — calls it against the account's default setup template.
-// Always creates (CL-6089) — never the `reuseExisting` land-hop path,
+// nowhere, and `instant-agent-create.ts` — THE one creation verb —
+// calls it against the account's default setup template.
+// Always creates — never the `reuseExisting` land-hop path,
 // which is `default-agent-workbench.ts`'s own call, not this one.
 
 import { createWorkbench } from "@corbits/chat-ui";

apps/web/src/agent-dm-launch.ts

@@ -1,5 +1,5 @@
 // The one path from an agent definition's id to "the person is in their
-// direct chat with it" (CL-6253) — the sidebar's agent rows are the one
+// direct chat with it" — the sidebar's agent rows are the one
 // caller. Mirrors `agent-chat-launch.ts`'s shape exactly, but through
 // `openAgentDm` (`kind: "chat"`, `reuseExisting: true`) rather than
 // `createWorkbench` directly: the first click mints the DM, every later

apps/web/src/agents-api.ts

@@ -119,7 +119,7 @@ export function listAgentInstances(tenantId: string): Promise<readonly AgentInst
  * native `GET /workflows/runs` listing's own predicate (`address IS NOT
  * NULL AND anchorRunId = id`), the same one `isTopLevelRun` uses. Used
  * wherever a page needs "real deployments only" — the Agent Directory.
- * Single-tenant (CL-8087 accepted loss): the deleted route expanded the
+ * Single-tenant (accepted loss): the deleted route expanded the
  * requested tenant to its whole descendant subtree via
  * `getDescendantTenants`; the native listing filters one tenant, so a
  * workspace parent sees only its own runs, not its child workbenches'.
@@ -135,7 +135,7 @@ export function listTopLevelRuns(tenantId: string): Promise<readonly AgentInstan
  * bare-array discovery route at `/models` (`ModelInfo[]`) — those are
  * different wire shapes. Disabled rows are filtered out here because the
  * catalog may retain them. Embedding-named models, Hugging Face Hub paths,
- * and bare `.gguf` names are also omitted (CL-6744) — this endpoint carries
+ * and bare `.gguf` names are also omitted — this endpoint carries
  * no offering capability lists, so the name-only
  * {@link isChatPickerModelName} gate is the available signal. */
 export function listCatalogModels(tenantId: string): Promise<readonly CatalogModel[]> {
@@ -160,7 +160,7 @@ export type AgentDefinitionDraft = typeof AgentDefinitionDraftResponse.infer.dra
  * Asks Myra to draft a starting system prompt (and optionally a
  * refined description, a model pick, and skills) from a name and a
  * plain-language purpose — the create-agent panel's "Create & chat"
- * flow (CL-6074). Hits `@corbits/agent-directory`'s
+ * flow. Hits `@corbits/agent-directory`'s
  * `POST .../planner/agent-definitions/draft`; never deploys anything
  * itself. A caller that gets a rejected promise here (Myra unavailable,
  * the draft timing out, an unparseable or out-of-inventory reply, or a

apps/web/src/agents-directory.ts

@@ -18,7 +18,7 @@ import {
 
 import type { AgentDefinition, AgentInstance } from "./agents-api";
 
-/** An `AgentDefinition` with its display name derived (CL-6413) — its own
+/** An `AgentDefinition` with its display name derived — its own
  * description when the definition was created with one, otherwise a
  * humanized reading of its immutable `name` slug. `name` itself stays the
  * slug throughout; nothing here mutates it. */
@@ -41,7 +41,7 @@ export function purposeAgentDefinitions(
  * (see `./agents-api.ts`), which excludes every non-top-level run
  * (workbench host, invited agent, task) server-side — the native
  * `GET /workflows/runs` listing's own predicate. That listing is
- * single-tenant (CL-8087 accepted loss: no descendant-subtree rollup),
+ * single-tenant (accepted loss: no descendant-subtree rollup),
  * so this still applies the shared name-based workbench-host filter as
  * defense in depth.
  */

apps/web/src/api.ts

@@ -53,8 +53,6 @@ export const ArtifactRowSchema = ArtifactListItemSchema.merge(
     content: "string",
   }),
 );
-// GET /api/tenants/:id/artifacts/:artifactId wraps the row in an `artifact`
-// envelope key (`@corbits/artifacts`' `mountArtifacts`, CL-8188).
 export const ArtifactDetailSchema = type({
   artifact: ArtifactRowSchema,
 });
@@ -179,7 +177,7 @@ export function rejectApproval(
 }
 
 /**
- * Sandboxed HTML preview URL for a Library artifact (CL-5879) — the same
+ * Sandboxed HTML preview URL for a Library artifact — the same
  * path an `<iframe sandbox>` in the canvas or Library detail pane loads,
  * and the "Open in new tab" affordance's `href`. Server-side (`GET
  * .../artifacts/:id/preview` in `@corbits/artifacts-hub`) answers 415 for

apps/web/src/app-error-boundary.tsx

@@ -1,4 +1,4 @@
-// The last line of defence (CL-6381): a render error anywhere in the tree
+// The last line of defence: a render error anywhere in the tree
 // used to leave the reader staring at a blank white page. React only offers
 // this as a class component (no hook equivalent exists), so it's the one
 // class in an otherwise function-component codebase.

apps/web/src/app.css

@@ -93,8 +93,8 @@ select:disabled,
   height: 100dvh;
 }
 
-/* A quoteable reference id under the error boundary's failure message
-   (CL-6632) — small and muted, same treatment as `.onboarding-error-refid`. */
+/* A quoteable reference id under the error boundary's failure message —
+   small and muted, same treatment as `.onboarding-error-refid`. */
 .app-boot-frame-refid {
   font-size: 0.75rem;
   color: var(--muted-foreground);
@@ -304,10 +304,10 @@ select:disabled,
   flex-direction: column;
 }
 
-/* CL-6092: the guided-fix banner, above the stage's own top bar — quiet
+/* the guided-fix banner, above the stage's own top bar — quiet
    until there is something to act on, then a clear (not alarming) nudge
    toward the fix. */
-/* Mount/unmount collapse (CL-6092): a grid-rows 0fr/1fr toggle so the
+/* Mount/unmount collapse: a grid-rows 0fr/1fr toggle so the
    banner's entrance/exit eases the stage's height rather than snapping it
    open or shut. `theme.css` (react-ui) already collapses every
    transition duration under `prefers-reduced-motion`, so this needs no
@@ -357,9 +357,9 @@ select:disabled,
   color: var(--card-foreground);
 }
 
-/* The Plugins gallery's own "couldn't find that connection" notice
-   (CL-6092): a deep link from the shell banner that names a provider the
-   gallery has no card for. */
+/* The Plugins gallery's own "couldn't find that connection" notice: a deep
+   link from the shell banner that names a provider the gallery has no
+   card for. */
 .plugins-connect-notice {
   display: flex;
   align-items: center;
@@ -1153,7 +1153,7 @@ select:disabled,
   overflow: auto;
 }
 
-/* Co-viewer cursors (CL-5958): positioned in the pane's own fractional
+/* Co-viewer cursors: positioned in the pane's own fractional
    coordinate space (see `PresenceCursor` in canvas-column.tsx), never
    intercepting pointer events — they are a read-only presence signal, not
    an interactive layer. */
@@ -1200,7 +1200,7 @@ select:disabled,
    that like any other. */
 
 /* Sidebar header: the brand mark, small and quiet — the mark is the brand
-   moment now (CL-6089 dropped the orange-rail era), not the switcher this
+ moment now (dropped the orange-rail era), not the switcher this
    dock replaced. */
 .shell-sidebar-brand-mark {
   display: flex;
@@ -1613,7 +1613,7 @@ select:disabled,
   white-space: nowrap;
 }
 
-/* Persistent pin glyph on pinned sidebar rows (CL-6657) — pinning must be
+/* Persistent pin glyph on pinned sidebar rows — pinning must be
    visible in the list itself, not only as a toast / settings bit. */
 .shell-ch-pin {
   flex-shrink: 0;
@@ -2269,8 +2269,8 @@ select:disabled,
   color: var(--muted-foreground);
 }
 
-/* A quoteable reference id under a consumer-language failure message
-   (CL-6360) — small and muted, never the focal point of the card. */
+/* A quoteable reference id under a consumer-language failure message —
+   small and muted, never the focal point of the card. */
 .onboarding-error-refid {
   font-size: 0.75rem;
   color: var(--muted-foreground);
@@ -2366,7 +2366,7 @@ select:disabled,
   min-height: 0;
 }
 
-/* Route-level Suspense fallback (CL-6370): centers the shared warm loader
+/* Route-level Suspense fallback: centers the shared warm loader
    instead of leaving a bare, un-laid-out box while a route chunk loads. */
 .shell-route-loading {
   display: flex;
@@ -3077,7 +3077,7 @@ tr.insights-row-clickable:hover {
   gap: 1rem;
 }
 
-/* CL-6124: the in-shell first-workbench screen — a chat, not a form. One
+/* the in-shell first-workbench screen — a chat, not a form. One
    prompt box, centered both ways in the stage; the stage owns the full
    width, no decorative panel. */
 .describe-first-workbench {
@@ -3124,7 +3124,7 @@ tr.insights-row-clickable:hover {
   color: var(--muted-foreground);
 }
 
-/* CL-6224: per-workbench Timeline — one wall-clock spine (chat messages,
+/* per-workbench Timeline — one wall-clock spine (chat messages,
    thread forks, routine runs, tasks, approvals), day dividers, and a right
    rail of per-day KPI counts. Matches the Insights stage's own visual
    language (hairline dividers, uppercase section labels, --shell-accent
@@ -3348,7 +3348,7 @@ tr.insights-row-clickable:hover {
   font-weight: 550;
 }
 
-/* ---------------- new-workbench picker (CL-6342) ---------------- */
+/* ---------------- new-workbench picker ---------------- */
 /* Screen 1 of the approved mock: one row list, no card grid. The
    selection rail (`box-shadow: inset 2px 0 0 0 var(--shell-accent)`)
    matches the sidebar's own active-row treatment exactly. */
@@ -3384,7 +3384,7 @@ tr.insights-row-clickable:hover {
   margin: 0;
 }
 
-/* The prompt box (CL-6628): the primary act on this screen, so it gets
+/* The prompt box: the primary act on this screen, so it gets
    the heaviest visual weight — a bordered, focus-ringed field with the
    send button inline, sized to read as "type here first". */
 .new-workbench-prompt {
@@ -3723,7 +3723,7 @@ tr.insights-row-clickable:hover {
   }
 }
 
-/* Mission Control (CL-6488/CL-6489) — stat strip, then a main column
+/* Mission Control — stat strip, then a main column
    (Needs you / In flight) beside a 320px rail (Jump back in / This week),
    the same "stack under 1100px" rule DESIGN.md gives every right rail. */
 .mission-control-layout {

apps/web/src/auto-workbench-title.test.ts

@@ -7,7 +7,7 @@ import {
   titleFromFirstMessage,
 } from "./auto-workbench-title";
 
-describe("titleFromFirstMessage (CL-6656)", () => {
+describe("titleFromFirstMessage", () => {
   test("trims and collapses whitespace into a single-line title", () => {
     expect(titleFromFirstMessage("  Help me\nplan  Q3  ")).toBe("Help me plan Q3");
   });
@@ -33,7 +33,7 @@ describe("titleFromFirstMessage (CL-6656)", () => {
   });
 });
 
-describe("autoNameFromFirstMessage (CL-6656)", () => {
+describe("autoNameFromFirstMessage", () => {
   test("names an ad-hoc New Workbench from the first message", () => {
     expect(autoNameFromFirstMessage(NEW_WORKBENCH_TITLE, "Plan the Q3 launch")).toBe(
       "Plan the Q3 launch",

apps/web/src/auto-workbench-title.ts

@@ -1,4 +1,4 @@
-// CL-6656: ad-hoc ("Just start talking") benches mint as "New Workbench" and
+// ad-hoc ("Just start talking") benches mint as "New Workbench" and
 // used to stay that way in the sidebar. Prefab templates already name the
 // room after the template title; this helper turns the person's first message
 // into a short sidebar title so blank benches get the same treatment once

apps/web/src/chat-artifact-open.ts

@@ -29,7 +29,7 @@ export function artifactContentFromDetail(
     title: detail.title,
     rendererKind,
     content: detail.content,
-    // Text-kind Library artifacts are co-editable (CL-5958 phase 2); the
+    // Text-kind Library artifacts are co-editable (phase 2); the
     // presence `/update` route's own write-grant check is the real gate
     // — this only decides which pane a capable viewer sees.
     canEdit: rendererKind === "doc",

apps/web/src/client-bootstrap.ts

@@ -44,7 +44,7 @@ function primaryTenantSlug(account: ClientBootstrapAccount): string {
   return base !== undefined && base.length > 0 ? base : `home-${account.id}`;
 }
 
-/** First-signup installer step (CL-8131): mints the account's primary
+/** First-signup installer step: mints the account's primary
  * tenant over the stock `POST /api/tenants` route (no `parentId`, so the
  * caller becomes its owner) when it does not already own one. This is
  * the "0→1" ruling's execution — never hub boot, never a CLI — and it