Editorial policy and page review standards
Last updated: September 29, 2026
PURPOSE
Built around specific tasks
Content is organized around tool purpose, usage steps, and result interpretation.
REVIEW
Checks scoped to changes
Coverage of tool tests, browser checks, and metadata audits varies by change.
CORRECTIONS
Errors are prioritized
Incorrect results, misleading explanations, privacy questions, and usage-boundary concerns move to the top of the queue.
QUALITY
Fill information gaps
Prioritize examples, scope, and failure cases that affect practical use.
1. Content and ownership
ToolsKit is independently developed and maintained by Evan. It publishes online tools, practical guides, and development articles. The maintainer is responsible for selecting content, revising pages, and handling issue reports.
The following principles guide new content and revisions: explain the purpose, then add relevant steps, inputs and outputs, scope, common errors, or references. Evidence and verification coverage depend on the conditions and limitations stated on each page.
2. Maintenance and review approach
- 01
Define the problem first
Clarify page purpose, target user, linked tools, and language versions.
- 02
Add executable information
Add relevant steps, examples, failure patterns, decision boundaries, fix actions, or review checklists.
- 03
Select checks and state their scope
Choose automated tests, browser checks, or reference review according to the change. Passing checks cover the tested conditions, not every tool, browser, or environment.
- 04
Schedule revisions from feedback
Prioritize confirmed issues by impact and reproducibility. An update date records a content change; it does not mean every feature was tested again.
3. Page quality criteria
Clear purpose
A user should understand quickly what problem the page is meant to solve.
Executable flow
The page should contain steps, input-output reasoning, or a troubleshooting route instead of only definitions.
Visible boundaries
Users should be able to see when the page applies, when it does not, and where common misreads happen.
Reviewability
Key pages should expose update metadata, correction paths, or other signs of active maintenance.
4. What prompts a revision
- A page has little beyond feature description and no longer supports independent task completion.
- Large sections repeat across pages without meaningful new scenarios, boundaries, or decisions.
- Search intent is clear, but the page still does not address the most common wrong inputs or failure cases.
- Tool behavior, external specifications, or dependencies change and the guidance no longer applies.
5. Improving incomplete content
Length alone does not make content useful. Revisions focus on concrete questions: what to enter, how to interpret the result, when the guidance does not apply, and what to check next.
- Guides should explain the actions and reasoning relevant to their topic, avoiding repetition of generic definitions.
- Tool-page revisions prioritize failure examples, usage scenarios, diagnostic logic, and result interpretation.
- Known defects need stated limitations or a planned fix. Pages that cannot offer useful help should be considered for merging or removal.
6. Boundaries for AI-assisted drafting
AI may assist with drafts, code, structure, or translation; the maintainer remains responsible for publication. Review focuses on technical facts, references, example results, and English/Chinese wording. AI-generated conclusions are not verification evidence. Experience reports and reviews should distinguish retained records, demonstrations, and the author’s interpretation, with known limitations stated.
7. Corrections and page responsibility
Evan is responsible for these pages. Use the contact page or [email protected] to share the page URL, reproduction steps, and expected result, leaving out sensitive data. Incorrect results, misleading guidance, privacy explanations, and blocked actions take priority. Revisions depend on impact and reproducibility, with no fixed update cycle.