@rubytech/create-realagent-code 0.1.175 → 0.1.179

package/payload/platform/plugins/memory/skills/document-ingest/SKILL.md

@@ -112,7 +112,6 @@ You produce typed-section JSON yourself from the cached text and the loaded onto
     {
       "kind": "<from closed enumeration>",
       "title": "<short title>",
-      "body": "<documentText.slice(sourceStart, sourceEnd)>",
       "summary": "<≤500 chars>",
       "sourceStart": <int>,
       "sourceEnd": <int>,
@@ -127,7 +126,7 @@ You produce typed-section JSON yourself from the cached text and the loaded onto
 }
 ```
 
-Use the natural-edge map below to pick `anchorEdge` per kind. `body` is the verbatim slice of the cached text between `sourceStart` (inclusive) and `sourceEnd` (exclusive). `summary` ≤ 500 chars.
+Use the natural-edge map below to pick `anchorEdge` per kind. The writer derives `:Section.body` from `cachedText.slice(sourceStart, sourceEnd)` — you emit offsets, not body text. Any `body` field you supply is logged and ignored. `summary` ≤ 500 chars.
 
 ```
 Section kind                     | Anchor edge (direction = from-anchor unless noted)
@@ -148,7 +147,7 @@ Other (label fallback)           | anchorEdge: null — section-kind closure mis
 
 `orphanCandidates` is the list of entities you wanted to emit but could not natural-edge. Surface them so the ontology can grow.
 
-**Chat mode.** Same top-level shape with every `sections[].kind = "Conversation"` and per-chunk properties: `summary`, `keywords`, `firstMessageAt`, `lastMessageAt`, `participantNames`, `messageCount`. Cover every message in chronological order with no gaps. Drop `anchorEdge` (always null) and `related`. Use the participant elementIds from § Participant confirmation; do not invent new ones.
+**Chat mode.** Same top-level shape with every `sections[].kind = "Conversation"`. `summary` is the per-section top-level field (already required by the schema) — the writer reads it from there and stamps `:Section.summary`. The `properties` bag is for non-authoritative per-chunk metadata only: `keywords`, `firstMessageAt`, `lastMessageAt`, `participantNames`, `messageCount`. The writer strips reserved writer-owned keys (`summary`, `title`, `body`, `bodyPreview`, `embedding`, provenance fields, etc.) from `properties` and logs the attempt — so duplicating an authoritative field into `properties` is a noisy no-op. Cover every message in chronological order with no gaps. Drop `anchorEdge` (always null) and `related`. Use the participant elementIds from § Participant confirmation; do not invent new ones.
 
 **Chunking — long inputs.** For inputs > ~525K chars, work through the cached text in deterministic ~525K-char windows with ~17.5K-char overlap, emit sections per window, and merge same-kind boundary straddlers yourself before calling `memory-ingest`. Inputs > 682K chars in chat mode are too large to keep within one turn — abort with `cause:"input-too-large"` and ask the operator to sessionize the export before re-dispatch.
 
@@ -247,6 +246,16 @@ The operator always sees what was attempted and where it stopped.
 
 For URL-based ingestion, run `memory-ingest-web` first with the URL and the WebFetch result. It writes the content to a temp file and routes through `memory-ingest-extract`, returning a fresh `attachmentId`. The rest of the pipeline (classify + ingest) is identical to a file upload. Pass `sourceUrl` and `sourceType: 'web'` to `memory-ingest`.
 
+**Page title is mandatory input — parse it before the call.** `KnowledgeDocument.name` is what shows up everywhere the document is referenced (`pickShortLabel` / `pickDisplayName` check `name` first), so it must be an editorial title, never body prose. Before calling `memory-ingest-web`, scan the WebFetch result for the page's title and pass it as the `title` parameter:
+
+- The readability output often starts with the page's `<title>` or `og:title` as a literal H1 (e.g. `# I am no longer the Official Representative for Reform UK`). Use that line verbatim — strip the leading `# ` and any trailing site name suffix (e.g. ` - Substack`, ` | Medium`).
+- When the readability output strips the title entirely (rare; some publishers prefix the body with marketing chrome), look at the URL's final path segment — `…/i-am-no-longer-the-official-representative` is the slug the publisher chose for the article.
+- Never invent a title from the article's opening paragraph. Body prose is exactly what this parameter exists to avoid.
+
+The server's fallback chain when `title` is absent: markdown H1 in the cached text → URL-derived slug → the URL itself. The librarian's job is to make the `caller` branch the common case; the server-side fallbacks are last-resort, not the default path. The response includes `titleSource` (`caller` / `h1` / `url-slug` / `url`) so the operator can see which path fired.
+
+**WebFetch limitation — subscriber-gated and JS-rendered pages (Substack, Medium-locked, paywalled news, gated PDFs, single-page apps).** `WebFetch` retrieves only the public, server-rendered surface. For gated articles the return is the meta-description / preview blurb / third-person summary that the publisher exposes to scrapers — not the author's original prose. Verbatim text is unrecoverable through this path. **Detect this before ingesting:** if the WebFetch result looks like a third-person synopsis of an article whose author would have written in first person, or carries publisher-summary framing (`# <Title> - Full Article Text` headers, marketing blurbs, "Subscribe to read"), abort the ingest and surface the gap to the dispatching agent in one line: *"WebFetch for `<url>` returned a publisher summary, not the article body. Ask the operator for a different input shape — for a Substack URL on the operator's own publication, use the `substack-import` plugin's export-archive flow (Substack admin → 'Export your data' → drop the ZIP); otherwise paste the article text into chat or attach a PDF / .txt export."* The dispatching agent relays this to the operator and re-dispatches with the pasted text, uploaded file, or Substack archive as the input. Do NOT proceed to `memory-ingest-web` with the summary — `:Section.body` will faithfully reproduce that summary downstream, defeating the verbatim-prose contract.
+
 ## Attachment metadata format
 
 Your task description contains attachment metadata in this format:

package/payload/platform/plugins/substack-import/skills/substack-import/SKILL.md

@@ -179,4 +179,5 @@ Audit query: `MATCH ()-[r]->() WHERE r.createdByAgent='substack-import' RETURN D
 - **Comments.** Substack export does not include reader comments.
 - **Inbound newsletter ingestion.** A Substack post landing in the operator's Gmail goes through `conversation-archive --source email` (Task 426), not this plugin.
 - **Live API or RSS poll.** Operator-dropped export only.
+- **Single-URL ingestion via `WebFetch`.** A Substack URL passed to `document-ingest` via `memory-ingest-web` will NOT recover the article's verbatim prose for gated / paywalled / subscriber-only posts — `WebFetch` retrieves only the public, server-rendered surface (meta-description, preview blurb, third-person publisher summary). Even for free public posts, JS-rendering on Substack means the WebFetch return is unreliable for verbatim capture. **If the operator hands over a Substack URL, advise:** (a) for the operator's own publication — run the export-archive flow this skill owns (Substack admin → "Export your data" → drop the ZIP); (b) for a third-party publication — paste the article text into chat, or save the page as PDF / `.txt` and attach the file. Then re-dispatch the librarian on the pasted text or attached file. Do not attempt URL-based ingestion as a fallback — the prose will be lost. (Recurring failure surfaced on session `d725df52` 2026-05-27; see `.tasks/465-memory-ingest-body-server-sliced-from-offsets.md` and the WebFetch-limitation note in `document-ingest/SKILL.md`.)
 - **Audio / video enclosures.** Substack podcast / video posts use additional file shapes the v1 plugin does not handle; only images referenced from post-body `<img>` tags are enclosed.

package/payload/premium-plugins/real-agent/plugins/brochures/skills/property-brochure/SKILL.md

@@ -95,7 +95,7 @@ If the operator did **not** request socials, do not run the step; adding 1 to 3
 
 ## Step 4 — populate, render, deliver
 
-**Before writing any copy, call `voice-retrieve-conditioning`** with `format: "long"` and `topic` set to the property's `addressLocality` or the marketing angle (e.g. "period townhouse in Hackney"). When the tool returns a non-null `styleCard`, condition the long-form narrative copy on it — the style card + exemplars feed the editorial copy register; brand-required structure rules still override. When the tool returns `{styleCard: null, exemplars: []}` (no operator voice profile yet, or `writer-craft` not enabled), fall back to the unconditioned register documented in `references/copy.md`. Emit narration `[property-brochure] voice-mirror styleCardBytes=<n> exemplars=<k>` when conditioning was applied.
+**Before writing any copy, call `voice-retrieve-conditioning`** with `format: "article"`, `length: "long"`, and `topic` set to the property's `addressLocality` or the marketing angle (e.g. "period townhouse in Hackney"). When the tool returns a non-null `styleCard`, condition the long-form narrative copy on it — the style card + exemplars feed the editorial copy register; brand-required structure rules still override. When the tool returns `{styleCard: null, exemplars: []}` (no operator voice profile yet, or `writer-craft` not enabled), fall back to the unconditioned register documented in `references/copy.md`. Emit narration `[property-brochure] voice-mirror styleCardBytes=<n> exemplars=<k>` when conditioning was applied.
 
 Read the seller brief once before writing any copy. Then walk the `{{ placeholder }}` tokens in both files top-to-bottom, substituting from the brief and `property.json`. The complete substitution contract — every token, its source field, its expected format and voice — lives in [references/placeholders.md](references/placeholders.md). Each token also has an inline `<!-- REPLACE: ... -->` comment in the template giving the short version.