technical-writing（技术写作）

Technical Writing

Use this skill when the main job is writing or restructuring internal technical documentation for builders and operators.

technical-writing is the documentation-cluster anchor for:

• technical specs
• architecture docs
• ADRs / decision records
• runbooks and incident procedures
• rollout / rollback / migration docs
• internal developer guides tied to implementation or operations

Read these support docs before choosing the mode or boundary:

• references/document-modes-and-boundaries.md
• references/quality-checklists.md
• references/docs-as-code-and-maintenance.md

When to use this skill

• A team needs a technical spec before implementation starts
• An engineer needs an architecture document or ADR that records trade-offs and decisions
• Operations needs a runbook, rollback guide, or incident response procedure
• A migration, rollout, or deprecation plan needs a durable written path
• A developer-facing implementation guide or internal onboarding doc needs to explain how a system works and how to work on it safely

When not to use this skill

• The main job is a published API portal, OpenAPI reference, SDK docs, or developer portal content → use api-documentation
• The main job is onboarding end users, tutorials, FAQs, screenshots, or help-center flows → use user-guide-writing
• The main job is release-note hygiene or semantic version / changelog upkeep → use changelog-maintenance
• The main job is product positioning, marketing copy, or GTM messaging → use marketing-automation or another marketing skill
• The main job is writing a plan for the work itself rather than the technical document artifact → use task-planning, api-design, or the relevant planning skill first

Instructions

Step 1: Classify the document mode

Normalize the request into one primary mode before drafting.

technical_writing_mode:
  primary_mode: spec | architecture | adr | runbook | migration | internal-guide
  audience: engineers | operators | mixed | unknown
  source_of_truth: repo | incident-notes | existing-doc | mixed | unknown
  lifecycle_state: draft | review | rewrite | maintenance
  docs_surface: markdown-repo | docs-site | wiki | unknown
  review_need: decision-signoff | operational-accuracy | handoff-clarity | unknown

Choose one primary mode per run:

• spec → planned change, scope, goals, non-goals, design, rollout
• architecture → system structure, components, interfaces, trade-offs
• adr → one material decision with options and rationale
• runbook → diagnose / operate / recover / escalate
• migration → move from old to new safely with rollback / compatibility notes
• internal-guide → developer-facing implementation or maintenance guidance

Step 2: Confirm the real audience and neighboring skills

Before writing, answer three questions:

1. Who will act on this document?
2. What decision or action should it enable?
3. Which neighboring documentation skills should stay out of scope?

Quick route-out table:

Step 3: Gather the minimum technical evidence

Do not draft from vibes alone. Pull the smallest credible evidence set first:

• current behavior or architecture notes
• interfaces / contracts / schemas / commands
• rollout or operational constraints
• known failure modes and recovery steps
• open questions or unresolved trade-offs

If evidence is missing, label assumptions explicitly instead of pretending the document is authoritative.

Step 4: Draft using the smallest mode-specific structure

Use only the sections that fit the chosen mode.

A. Technical spec

Use for planned work before implementation.

Recommended skeleton:

# <Feature / Change> Technical Specification

## Overview
## Problem
## Goals
## Non-goals
## Constraints
## Proposed design
## Interfaces / data / dependencies
## Risks and mitigations
## Rollout and rollback
## Open questions

B. Architecture document

Use for system structure and longer-lived technical explanation.

Recommended skeleton:

# <System> Architecture

## Context
## Responsibilities and boundaries
## Components
## Data / request flow
## Key decisions
## Failure modes
## Security and performance notes
## Operational considerations
## Future changes / known limits

C. ADR

Use when one decision needs a durable record.

Recommended skeleton:

# ADR: <Decision title>

- Status:
- Date:
- Owners:

## Context
## Decision
## Alternatives considered
## Consequences
## Follow-up actions

D. Runbook

Use when someone will operate, diagnose, or recover a system.

Recommended skeleton:

# <Service> Runbook

## Purpose
## Preconditions / access
## Signals and symptoms
## Immediate checks
## Standard operating procedure
## Escalation path
## Rollback / recovery
## References

E. Migration / rollout guide

Use when systems or users move from old to new.

Recommended skeleton:

# <Migration> Guide

## Scope
## Preconditions
## Compatibility / breaking changes
## Step-by-step migration
## Validation
## Rollback
## Communication notes

F. Internal developer guide

Use for implementation-facing reference that is not end-user help.

Recommended skeleton:

# <Topic> Developer Guide

## What this system does
## When to use / not use it
## Key concepts
## Local development or operational workflow
## Common pitfalls
## Troubleshooting / escalation
## Related docs

Step 5: Apply document-type rules instead of one generic style

Use these rules aggressively:

• Specs should separate goals from non-goals.
• Architecture docs should explain boundaries and trade-offs, not every code path.
• ADRs should capture a decision, not become a full design doc.
• Runbooks should optimize for fast action under pressure.
• Migration guides should foreground compatibility, order of operations, and rollback.
• Internal guides should explain operational or implementation reality, not market value props.

Step 6: Make the document docs-as-code friendly

Default to repo-friendly, reviewable writing:

• stable headings
• concise bullet lists where operators scan
• explicit commands, paths, owners, and prerequisites
• dated decisions and status for ADR-like docs
• links to source-of-truth docs instead of duplicated narrative when possible

If the doc belongs in a docs site or wiki later, write the technical core first and let the publishing surface adapt it.

Step 7: Run the review checklist

Before finalizing, verify:

1. The audience is named or obvious.
2. The document tells the reader what decision or action it enables.
3. Assumptions and unknowns are labeled.
4. Commands / interfaces / rollback steps are concrete where relevant.
5. Neighboring documentation skills are not being absorbed.
6. The title and section layout match the chosen mode.

Step 8: Return one of these output shapes

Preferred output:

# Technical Writing Brief

## Mode
- Primary mode:
- Why it fits:
- Audience:

## Source material used
- Repo/docs/evidence:
- Assumptions / gaps:

## Draft structure
1. section
2. section
3. section

## Writing notes
- Key decisions / actions enabled:
- Risks / unknowns:

## Route-outs
- Neighboring skills before/after this work:

Or, if the user asked for the finished artifact, produce the chosen document mode directly with the mode-specific structure above.

Examples

Example 1: Internal design doc before implementation

Input

Write a technical spec for moving our worker queue from Redis lists to Redis streams. Engineers need goals, constraints, rollout, and rollback before coding.

Good output direction

• mode: spec
• audience: engineers
• include goals, non-goals, constraints, design, rollout, rollback, open questions
• keep API portal publishing out of scope

Example 2: Architecture decision capture

Input

We chose Postgres logical replication over dual writes. Record the decision and alternatives in an ADR.

Good output direction

• mode: adr
• capture context, decision, alternatives, consequences, follow-up
• keep the document short and decision-focused

Example 3: Incident runbook

Input

Write a runbook for when the payments worker backlog spikes and retries start timing out.

Good output direction

• mode: runbook
• include symptoms, immediate checks, operating steps, escalation, rollback / recovery
• optimize for operator speed, not essay-style explanation

Example 4: Boundary with user docs

Input

Create a beginner tutorial with screenshots that teaches customers how to export their data.

Good output direction

• route to user-guide-writing
• explain that this request is end-user onboarding / help content, not internal technical documentation

Best practices

1. Choose the document mode before writing the body.
2. Keep internal technical docs decision- and action-oriented.
3. Write only the sections the mode needs; do not force every template into every document.
4. Separate internal design / ops docs from API portals, user help, and release notes.
5. Prefer docs-as-code structure: reviewable Markdown, stable headings, and source-linked facts.
6. Label assumptions and unresolved questions explicitly.
7. For runbooks and migrations, make rollback and escalation easy to find.

References

• Diátaxis
• Write the Docs — Docs as Code
• Write the Docs — How to write software documentation
• Architectural Decision Records
• Keep a Changelog