Product Scout — Cycle Report 2026-07-13

_Trigger: doc-update on Great product features in the wild: Canva Pause Subscription._

Summary

The Product Backlog was concurrently updated by multiple parallel cycles during this trigger. The backlog now contains 41 opportunities. This document captures my unique findings from this cycle that should be merged into the Product Backlog.

---

New Opportunity: @shm/block-fragment — Block Fragment Annotation Card Package

Problem

Text-selection comments in Seed have two competing presentation modes: full-block rendering (too tall for comment threads) and compact fragment display (not yet built). The Block Fragment Annotation Card design doc states: "The full-block rendering is useful for context, but it is too tall and visually heavy for comment threads, replies, and embedded discussions."

Evidence

Block Fragment Annotation Card — Complete design spec with problem, visual mockups, implementation scope (1-2 days), rabbit holes, no-gos, and meeting notes

Discussions Panel Is Confusing — Confirms thread UI is broken

What I want for the commenting experience point #2 — Requests comment blocks behaving like document blocks with block-range links

Rich Text Editing Inspirations — Competitive UX analysis of Coda, Slite, Linear, Google Docs margin-based annotation patterns

create-comments story — Active design conversation confirming vertical action strip direction

Users

Seed frontend developers building the commenting UI; any team building annotation/comment systems on block-based editors (Plate, TipTap, ProseMirror).

Solution

A reusable React component package @shm/block-fragment that renders block fragment annotations: compact fragment card for selected text ranges as primary content, full-block fallback for non-fragment embeds, rich formatting preservation, Unicode-safe codepoint offset extraction, optional expand/collapse context affordance.

MVP

BlockFragment UI component + text range extractor (Unicode codepoint offsets with annotation shifting) + integration points. Already scoped at 1-2 development days.

Distribution

npm install @shm/block-fragment. Integrated into Seed commenting pipeline. Blog post. XState/Plate communities.

Pricing

Free OSS (MIT). Consulting: custom fragment renderers ($3k–$8k).

Competition

No direct OSS competitor. ProseMirror decorations (not extractable), TipTap bubble menu (toolbar, not card). Notion/Coda margin threads (proprietary).

Confidence

9/10 — Fully designed with visual mockups, scope, rabbit holes, no-gos. 1-2 day build. Directly supports Opp 28 and Opp 29. Reuses existing Seed rendering stack.

First experiment

Build BlockFragment component in isolation. Test with 3 scenarios: simple text range, range with inline formatting, invalid/unsupported range. If all pass → publish.

Roadmap

V0 (week 1): Component extraction + basic fragment rendering. Comment integration.

V1 (week 2): Block-fragment embed integration. Rich formatting preservation.

V2 (month 1): Expand/collapse affordance. Storybook. Blog post.

V3 (quarter 2): Multi-block selection support.

---

Evidence Updates for Existing Opportunities

Opp 28: Comment Threading State Machine (@shm/thread-machine)

New evidence this cycle:

The create-comments story is under active design with acceptance criteria

Block Fragment Annotation Card provides a complementary component for rendering selected text fragments

Horacio posted an active design comment proposing to consolidate comment/link actions into a single vertical action strip

Confidence upgrade: 8/10 → 9/10

Opp 29: Seed Commenting UX Consolidation

New evidence this cycle:

5 explicit pain points documented in What I want for the commenting experience: full-window panel, block-range links, smooth mobile, mobile editor lag, web redirect

Discussions Panel Is Confusing — Threading counting bugs: "Reply (8)" shows 3 visible replies, hidden comment chains

Horacio posted active design direction on create-comments story

Confidence: 9/10 (active design direction from product lead)

Opp 34: Seed Pause Subscription Feature

New evidence this cycle:

Post updated with bidirectional links to feature request and backlog (cross-account validation)

Feature request notes: "Not important but something we can consider in the future" — low urgency, high delight

No confidence change (8/10) — validated but not time-sensitive

---

Recommended Actions

Merge Opp 38 (block-fragment) into Product Backlog as new opportunity (recommended rank: #2 — very low effort, fully designed)

Merge evidence updates for Opp 28 (confidence 9/10), Opp 29 (active design direction), Opp 34 (cross-account validation)

Update ranking to include block-fragment at #2 position

---

_Generated: 2026-07-13. Agent: Horacio's Product Scout_