Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

Doc Sync

ASecurity

Synchronize documentation with the current codebase. Fix drift between docs/prd.md, docs/system-design.md, and actual source code.

16 stars
0 votes
0 copies
0 views
Added 9/20/2026
ai-agentsbashrefactoringgitapidocumentation

Works with

claude codecliapi

Security Analysis

A100/100

Scanned 9/20/2026

Install to Claude Code

$npx -y skills add woditschka/agentic-coding-reference --skill doc-sync --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Doc Sync?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Doc Sync
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/woditschka-doc-sync-765a6031/badge)](https://www.skillsdirectory.com/skills/woditschka-doc-sync-765a6031)

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
SKILL.md
---
name: doc-sync
description: >-
  Synchronize documentation with the current codebase. Fix drift between
  docs/prd.md, docs/system-design.md, and actual source code.
compatibility:
  - claude-code
  - github-copilot
  - opencode
reads:
  - docs/prd.md
  - docs/system-design.md
metadata:
  version: "1.0"
  author: team
---

# Doc Sync

Synchronize `docs/prd.md` and `docs/system-design.md` with the current codebase. Fix drift, add missing items, remove stale references.

## When to Run

- After implementing features or refactoring code
- Before starting a new feature cycle
- Periodically to prevent documentation drift

## Instructions

### Phase 1: Explore Current Codebase

Build a complete picture of what is implemented — via a read-only exploration subagent where the tool offers one (Claude Code: Explore), else by reading directly:

1. Read all production source files (the roots declared in `scripts/layout.toml`) -- note every type, interface, function, field
2. Read the project's configuration files where they exist (the config example and config module the project defines)
3. Read all test files to understand tested behavior
4. Read all ADR files under `docs/adr/`

Capture: type names (exact casing), type and interface fields (exact types), public vs internal visibility, module dependencies, pipeline step ordering, CLI arguments, configuration properties.

### Phase 2: Diff Against Documentation

Read `docs/prd.md` and `docs/system-design.md`.

Compare the codebase snapshot against both documents. Identify:

**In PRD:**
- Features implemented but not documented (missing requirement IDs)
- Features documented but not implemented (stale requirements)
- Configuration properties that changed, were added, or were removed
- CLI arguments that changed
- Behavioral details that drifted (thresholds, defaults, fallback logic)

**In System Design:**
- Type names that changed (case matters)
- Type fields that were added, removed, or retyped
- Module structure changes (new modules, moved files, visibility changes)
- Pipeline step ordering drift
- Error handling changes
- Contracts-table entries out of sync with source (types or interfaces renamed, added, or removed)

### Phase 3: Update Documents

Apply all fixes. Follow these rules strictly:

**Document boundaries** (per the harness-project API roster; form rules in the `document-writing` skill):
- PRD = *what* the system does. No source code, function names, type names, or implementation constructs.
- System design = *how* it is built: contracts (purpose plus source pointer), module structure, pipeline, error handling. Source is authoritative for type and function definitions; never copy them.

**Writing standards** (from the `document-writing` skill):
- No prohibited words: "significant", "arguably", "might", "would help", "should result in"
- No "some", "many", "most" without percentages
- No vague adjectives without data
- 70% of sentences under 20 words, max 30 words
- Acronyms defined on first use
- One idea per sentence

**Preservation rules:**
- Keep existing requirement IDs stable
- Add new IDs at the end of their section
- Never renumber existing IDs (downstream references depend on them)

### Phase 4: Validate

Invoke the `doc-reviewer` agent with this preamble:

> You are a read-only reviewer. Inspect files read-only with your file-viewing and search tools. Only permitted Bash commands: `scripts/gate.sh verify`. Do not write code, scripts, or temporary files. Never use system `/tmp`; use `.scratch/tmp/` for any temporary output.

The reviewer validates against the `document-writing` skill's checklist:
1. Structural checks (cross-references, tables, code blocks)
2. Cross-document coherence (requirement IDs, config properties, type fields, constants)
3. Writing standards (prohibited words, sentence length, acronyms)
4. Document boundaries (PRD has no source code; system-design has no copied source)

### Phase 5: Fix Review Issues

Apply fixes for any `[AUTOFIX]` or `[BLOCKED]` issues the reviewer found. Re-run the reviewer if changes were substantial. Stop when the reviewer returns APPROVED.

## Maintenance Rules, Compaction, Format Migration

The stack-independent half of this skill lives in [`maintenance.md`](maintenance.md) beside this file: the Maintenance Rules table, the Compaction procedure, and the Format Migration procedure. Read it whenever the task is maintenance, compaction, or migration; the phases above stay stack-bound, and the Instructions above name this stack's Contracts-table constructs.

## Output

Report a summary of changes made:
- Lines added/removed/changed per document
- New requirement IDs added
- Stale items removed
- Review result (APPROVED or remaining issues)

Attribution

woditschkawoditschka
View sourceMore from woditschka →
SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Related Skills

Caveman

Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.

1023331 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

686011 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3331 votes

catchup

Recovers prior coding-agent session context by running `catchup <agent> --since-compact`, which extracts a clean summary of a previous Codex, Claude Code, Antigravity, OpenCode, or Pi Agent session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", or asks to recover/summarize a previous session before continuing. Do NOT use for the current conversation, git history, or any non-agent log.

611 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →