document

Generates five markdown documentation files from project context, epics, stories, and ADRs.

Updated May 25, 2026
One-click install
npx skills add https://github.com/DATA-AI-XYZ/Tandem --skill document-data-ai-xyz
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: document
Source: https://github.com/DATA-AI-XYZ/Tandem/tree/main/skills/document
Command: npx skills add https://github.com/DATA-AI-XYZ/Tandem --skill document-data-ai-xyz

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Project knowledge accumulates across PM artefacts — context files, epics, features, stories, and ADRs — but there is no coherent, shareable documentation set synthesised from it. This Skill reads those accumulated sources and authors a default five-document markdown set into the documentation/ folder, eliminating a separate manual doc-writing pass. ## Core Features & Use Cases - Five-document authoring: Writes overview.md, getting-started.md, architecture.md, decisions.md, and features.md into documentation/, each with a dedicated authoring guide. - Source-grounded content: Reads PROJECT-CONTEXT.md, epics, features, stories, ADRs, and the codebase, marking gaps as "not yet documented" instead of inventing facts. - Machine-checkable source anchors: Appends a mandatory anchor block (files, scripts, ADRs, commands) to each document and stamps generated_at via the wiki-anchor-stamp.js script so a board can verify document freshness. - Use Case: After several sprints of PM artefacts have accumulated, invoke /tandem:document to produce a complete project wiki — overview, setup guide, architecture, ADR digest, and feature reference — then render it with npm run pm:docs. ## Quick Start Ask the assistant to generate the project documentation set from the existing PM artefacts using the document skill.

Frequently Asked Questions about document

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

FAQPage Schema
How do I generate project documentation from existing markdown artefacts?

Invoke the document skill (or /tandem:document) after PM artefacts exist. It reads PROJECT-CONTEXT.md, epics, features, stories, ADRs, and the codebase, then writes five markdown files into the documentation/ folder in a single pass.

What documents does the Tandem document skill write?

It writes exactly five files: overview.md, getting-started.md, architecture.md, decisions.md, and features.md, all placed flat in the documentation/ folder. It never creates subfolders or touches any other file in that directory.

Will it overwrite documentation files I wrote by hand?

No. The skill only writes the five files in its default doc set. Any other markdown file in documentation/ is treated as hand-authored, left untouched, and listed in the final summary as not written by the skill.

What happens if a source file like PROJECT-CONTEXT.md is missing?

The skill treats a missing source as "not present", notes the gap inside the relevant document section as "not yet documented", and continues authoring. It never aborts or invents facts to fill the gap.

Does the document skill generate HTML output?

No. It authors markdown only. HTML rendering is a separate step handled by the pm:docs script (generate-docs.js), which you run afterwards with npm run pm:docs.

Why does each generated document need a source-anchor block?

The anchor block lists the files, scripts, and ADRs each document's claims rest on, letting the board verify freshness. Without it the document renders as unassessable; the wiki-anchor-stamp.js script sets generated_at from the body's last real change.