From we
Documentation CONTENT consistency check. Finds outdated code references, contradictions, semantic duplicates. Reads docs thematically, not file-by-file. Use when user mentions "consistency", "outdated docs", "contradictions", "/we:doc-check".
How this skill is triggered — by the user, by Claude, or both
Slash command
/we:doc-checkThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
You are an architect reading documentation. Your brain finds inconsistencies, not regex.
You are an architect reading documentation. Your brain finds inconsistencies, not regex.
/we:doc-check # Full check (all themes)
/we:doc-check Memory # Only one theme
/we:doc-check "Auth System" # Quotes for multi-word themes
Not file-by-file. Collect ALL docs about ONE theme, read them together.
Theme Discovery: Scan the project's docs/ directory and identify themes from folder structure and file names. Common themes: Architecture, Auth, Database, API, Deployment, Testing.
Don't hold in memory. Write to findings file as you go:
## [Theme]
- [ ] `file.md:123` — What's wrong
Doc says: X
Code has: Y
CONFIDENCE: 95%
DECISION: Doc is outdated
FIX: Update code examples
Hierarchy: Code (runs) > ADR (decided) > Docs (explains)
| Level | Meaning | Action |
|---|---|---|
| 95-100% | Code + ADR agree | Fix immediately |
| 70-94% | Code works, no ADR | Probably doc outdated |
| 50-69% | Conflicting signals | Note in findings, review later |
| <50% | Unclear | ASK USER immediately |
| Tag | Meaning |
|---|---|
| CRITICAL | Following the doc causes errors |
| HIGH | Confusing, outdated |
| MEDIUM | Duplicate, redundant |
# Does the referenced path exist?
ls path/to/file.py
# Does the referenced function exist?
grep -n "def function_name" path/to/file.py
# When was it last changed?
git log -1 --format="%ai %s" path/to/file.py
/we:doc-review = FORMAT (frontmatter, paths, size) — automatable
/we:doc-check = CONTENT (contradictions, outdated) — needs judgment
Run /we:doc-check first (content), then /we:doc-review (format).
npx claudepluginhub weside-ai/claude-code-plugin --plugin weGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.