Changelog, dev startup, README, AgentHub 0.4.0, model catalog, and landing site (#7)

Branch-length batch covering tooling, the model layer, the Web App and the public
surfaces. Highlights:

- Changelog: a per-release `changelog/<version>/` tree, grouped by the surface each
  change touches, with a root CHANGELOG.md holding one line per release.
- Dev startup: `scripts/dev-prebuild.mjs` serializes the skills+core prebuild behind a
  lock and keeps `pnpm install` current; `pnpm dev` runs server+web together.
- AgentHub 0.3.3 -> 0.4.0: OmniMessage complete payloads carry one opaque `fidelity`
  object in place of item-level `signature`/`phase`, threaded verbatim through Trace,
  replay and resume; malformed classification adapted to the new error types.
- Model layer: a model is always referenced by an explicit `(provider, model_id)` pair.
  The provider is never inferred, guessed or defaulted -- both the catalog inference and
  the unique-match config resolution are gone, and CLI, SDK, server routes and
  run_subagent all require the complete pair. Catalog gains the Qwen Token Plan, Qwen
  Pay-As-You-Go and Fireworks AI gateways, plus an expanded OpenRouter group.
- Web App: catalog preset sync and per-group speed test on the Models page, positional
  slash commands, a markdown renderer, skill-library update reminders, and a vertically
  centred draft page whose upward menus size themselves to the room available.
- Public surfaces: restructured READMEs, the penguin.ooo landing site and blog, refreshed
  benchmark results for both suites, and the demo videos playing on the landing page.

Includes the fixes from a full review of the branch: 23 confirmed findings, among them a
provider-inference bug that could send one vendor's API key to another vendor's endpoint,
and an Escape handler that destroyed the composer's contents unrecoverably.

Verified on the branch head: pnpm test (1127 passing, 7 packages), pnpm typecheck and
pnpm format:check clean, Playwright e2e 14/14.

packages/core/src/session-title.ts

@@ -22,6 +22,31 @@ const EXCERPT_MAX_CHARS = 2000;
 /** Cap on title length (fallback truncation for when the model occasionally ignores the constraint). */
 const TITLE_MAX_CHARS = 30;
 
+/**
+ * Special message markers that must never leak into a title. These are machine-inserted
+ * XML-ish blocks (a skill invocation wraps the body in `<use_skills>…</use_skills>`, a
+ * subagent handoff / scheduled task prepend their own blocks) — meaningful to the runtime,
+ * noise in a title. The list is a fixed allowlist so ordinary angle-bracket text (e.g. a
+ * user pasting `<div>`) is left untouched.
+ */
+const MARKER_TAGS = ["use_skills", "handoff_from", "scheduled_task"];
+
+/**
+ * Strips machine-inserted marker blocks (see MARKER_TAGS) from conversation text so titles are
+ * built from the human-meaningful body only — both the material sent to the model and the
+ * fallback derived from the raw first message. Removes the paired `<tag>…</tag>` block (any
+ * inner content, across lines) and any stray unpaired `<tag>` / `</tag>` left behind.
+ */
+export function stripConversationMarkers(text: string): string {
+  let out = text;
+  for (const tag of MARKER_TAGS) {
+    out = out
+      .replace(new RegExp(`<${tag}>[\\s\\S]*?</${tag}>`, "g"), "")
+      .replace(new RegExp(`</?${tag}>`, "g"), "");
+  }
+  return out.trim();
+}
+
 export interface SessionTitleResult {
   /** The sanitized title; null when material is insufficient, the request fails, or the output is empty. */
   title: string | null;
@@ -44,6 +69,7 @@ export function buildTitlePrompt(userExcerpt: string, assistantExcerpt: string):
     "- Write the title in the SAME language the user is using.",
     "- Keep it short: at most 6 words, or ~16 characters for CJK.",
     "- Output ONLY the title text — no quotes, no trailing punctuation, no explanation.",
+    "- Answer immediately — do not think aloud or produce chain-of-thought.",
     "",
     "[User]",
     clip(userExcerpt),
@@ -51,12 +77,15 @@ export function buildTitlePrompt(userExcerpt: string, assistantExcerpt: string):
   if (assistantExcerpt.trim()) {
     lines.push("", "[Assistant]", clip(assistantExcerpt));
   }
+  // The trailing empty think block makes many reasoning models treat their thinking phase
+  // as already closed, so the one-off request spends its budget on the title itself.
+  lines.push("", "<think></think>");
   return lines.join("\n");
 }
 
-/** Sanitizes model output into a title: strips leading/trailing quotes/brackets and trailing punctuation (until stable), collapses whitespace, and truncates if too long; returns null for an empty result. */
+/** Sanitizes model output into a title: strips any leaked marker blocks and leading/trailing quotes/brackets and trailing punctuation (until stable), collapses whitespace, and truncates if too long; returns null for an empty result. */
 export function sanitizeTitle(raw: string): string | null {
-  let t = raw.replace(/\s+/g, " ").trim();
+  let t = stripConversationMarkers(raw).replace(/\s+/g, " ").trim();
   // Stripping quotes can expose more punctuation underneath (or vice versa), so strip repeatedly until stable.
   for (let prev = ""; prev !== t;) {
     prev = t;
@@ -81,10 +110,14 @@ export async function generateTitleWithLLM(
   llm: LLMInterface,
   args: { userText: string; assistantText: string; signal?: AbortSignal },
 ): Promise<SessionTitleResult> {
-  if (!args.userText.trim()) {
+  // Strip machine markers before the model ever sees the material, so a skill-invocation
+  // block (or handoff / scheduled-task marker) can't bleed into the generated title.
+  const userMaterial = stripConversationMarkers(args.userText);
+  const assistantMaterial = stripConversationMarkers(args.assistantText);
+  if (!userMaterial.trim()) {
     return { title: null, usage: null };
   }
-  const prompt = buildTitlePrompt(args.userText, args.assistantText);
+  const prompt = buildTitlePrompt(userMaterial, assistantMaterial);
   const gen = llm.streamGenerate({
     newMessages: [userText(prompt)],
     ...(args.signal ? { signal: args.signal } : {}),