bean-entry-handwritten

Stars1

Forks0

UpdatedApril 14, 2026 at 11:51

Create or update a bean entry from handwritten scans while preserving line breaks, handling CompanionComment formatting correctly, maintaining spans, and avoiding low-value boilerplate transcription.

Installation

Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.

Source

codekiln

codekiln/beans

View GitHub Repository View Creator Repositories

Download

Related occupationsSOC

Based on SOC occupation classification

Data Entry KeyersOffice and Administrative Support Occupations·SOC 43-9021

SKILL.md

readonly

name	bean-entry-handwritten
description	Create or update a bean entry from handwritten scans while preserving line breaks, handling CompanionComment formatting correctly, maintaining spans, and avoiding low-value boilerplate transcription.

Handwritten bean entry

Use this skill when a bean entry is created or revised from handwritten scans, notebook pages, or photographed brew notes.

Workflow

Import source images into public/images/ and place them near the related transcription in the bean body.
Do two transcription passes before finalizing, especially for place names, timings, and unusual wording.
Preserve the author's line breaks instead of normalizing into prose paragraphs.
In CompanionComment, use the from prop as the only attribution.
Do not add inline signatures such as - unicorn or - codekiln inside companion-comment bodies.
If exact companion-comment line breaks matter, use explicit <br /> tags inside the component body.
Skip repeated printed-label boilerplate when the scan already preserves it and the text is not central to the entry.
If the notes say a consumable span opened, closed, ran out, or was replaced, update the relevant file in src/content/spans/ in the same task.
Close completed spans with endDate, create successor spans when a new consumable replaces the old one, and link the transition from the bean body text.
If related content lookup reveals duplicate coffee pages for the same roast, consolidate to one canonical content entry rather than keeping parallel duplicates.

Output target

The markdown should read like a faithful transcription, not a paraphrase.
The scans should remain visible in the entry unless the user asks for text only.
Implied housekeeping should be completed, not merely noted.

More from this repository

same repository

bean-entry

codekiln/beans

Create a new bean entry with the correct file shape, public identifier, frontmatter, imported images, and Beans-specific markdown conventions.

2026-04-141

land-the-plane

codekiln/beans

Close out a Beans work session cleanly by validating work, finishing Beads-managed tasks correctly, landing on main, and pruning closed worktrees.

2026-04-141

playwright-preview

codekiln/beans

Capture local dev-server screenshots with Playwright for Beans preview workflows, especially for `/beans/` routes and visual verification of new pages or content edits.

2026-04-141

rulesync-preferences

codekiln/beans

Use when adding, changing, or reorganizing AI guidance in this repository. Prefer Rulesync skills over commands, subagents, or hooks for reusable workflows, and always edit Rulesync sources before generating tool-specific AI config files.

2026-04-141

rulesync

codekiln/beans

Generates and syncs AI rule configuration files (.cursorrules, CLAUDE.md, copilot-instructions.md) across 20+ coding tools from a single source. Use when syncing AI rules, running rulesync commands, importing or generating rule files, or managing shared AI coding configurations.

2026-04-141

term-entry

codekiln/beans

Create a new glossary term entry with the correct schema, compact body, and lightly sourced Beans field-guide tone.

2026-04-141

name	bean-entry-handwritten
description	Create or update a bean entry from handwritten scans while preserving line breaks, handling CompanionComment formatting correctly, maintaining spans, and avoiding low-value boilerplate transcription.

Handwritten bean entry

Use this skill when a bean entry is created or revised from handwritten scans, notebook pages, or photographed brew notes.

Workflow

Import source images into public/images/ and place them near the related transcription in the bean body.
Do two transcription passes before finalizing, especially for place names, timings, and unusual wording.
Preserve the author's line breaks instead of normalizing into prose paragraphs.
In CompanionComment, use the from prop as the only attribution.
Do not add inline signatures such as - unicorn or - codekiln inside companion-comment bodies.
If exact companion-comment line breaks matter, use explicit <br /> tags inside the component body.
Skip repeated printed-label boilerplate when the scan already preserves it and the text is not central to the entry.
If the notes say a consumable span opened, closed, ran out, or was replaced, update the relevant file in src/content/spans/ in the same task.
Close completed spans with endDate, create successor spans when a new consumable replaces the old one, and link the transition from the bean body text.
If related content lookup reveals duplicate coffee pages for the same roast, consolidate to one canonical content entry rather than keeping parallel duplicates.

Output target

The markdown should read like a faithful transcription, not a paraphrase.
The scans should remain visible in the entry unless the user asks for text only.
Implied housekeeping should be completed, not merely noted.