scribe

Generates shareable documents from CALM decisions, logs, and materials.

3|1|Updated Dec 9, 2025
One-click install
npx skills add https://github.com/isizono/calm --skill scribe-isizono
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: scribe
Source: https://github.com/isizono/calm/tree/main/skills/scribe
Command: npx skills add https://github.com/isizono/calm --skill scribe-isizono

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Long-running projects accumulate decisions, discussion logs, and materials inside CALM's database, but that knowledge stays locked in internal records. This Skill turns those records into standalone documents (ADRs, meeting minutes, design docs) that anyone can read without CALM access. ## Core Features & Use Cases - Source selection: Pick one or more completed activities as the source, with automatic collection of related decisions, logs, and materials via CALM tools. - Structure-first workflow: Presents an outline for user approval before writing, splits large volumes into multiple documents, and tracks each writing task as a [執筆] activity. - Reader-safe output: Writes self-contained Markdown with no internal IDs, marks gaps with <!-- TODO: 要確認 -->, and follows intent:document tag notes for style. - Use Case: After finishing a design discussion tracked in CALM, ask for an ADR and receive a Markdown document covering background, decisions, and rejected alternatives, ready to commit to your repository. ## Quick Start Ask the AI to turn the completed design discussion activity in CALM into an ADR document and confirm the proposed outline before writing.

Frequently Asked Questions about scribe

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate an ADR from CALM discussion records?▼

Invoke the skill with an activity ID or pick from completed activities. It collects decisions, logs, and materials for the linked topics, proposes an outline for approval, then writes a self-contained Markdown ADR with background, decisions, and rejected alternatives.

What sources does the scribe skill use to write documents?▼

It gathers decisions via get_decisions, discussion history via get_logs, and stored materials via search and get_material, all scoped to the topic IDs of the source activity. It can also search by tags to pull in related topics supplementally.

Can I combine multiple CALM activities into one document?▼

Yes. The skill asks whether to merge multiple activities into a single document during source selection. Conversely, if the collected material is too large, it proposes a split and creates a separate writing activity per document.

Does the generated document reference CALM internal IDs?▼

No. Documents are written for readers without CALM access, so internal IDs like decision numbers are never referenced; content is described directly. Cross-document references use file names, and gaps are marked with TODO comments.

When should I not use the scribe skill?▼

Do not use it for reorganizing or consolidating CALM context itself (recompose-context) or extracting lessons learned (postmortem); those are separate skills. Scribe only produces external-facing documents from existing records.