skills.rest
dkyazzentwatwa
dkyazzentwatwaCOMMUNITY
Explore all skills by this author →
GitHub

output-reviewer

Reviews AI-generated drafts for accuracy, PII exposure, tone, and readiness before publishing.

2|1|Updated Apr 26, 2026
Search Tags:#content review#review#privacy#proofreading#fact check#pii detection#draft qa
One-click install
npx skills add https://github.com/dkyazzentwatwa/skill_starter_pack --skill output-reviewer-dkyazzentwatwa
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: output-reviewer
Source: https://github.com/dkyazzentwatwa/skill_starter_pack/tree/main/output-reviewer
Command: npx skills add https://github.com/dkyazzentwatwa/skill_starter_pack --skill output-reviewer-dkyazzentwatwa

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Sending or publishing an AI-generated draft without review risks factual errors, leaked personal data, unsupported claims, and off-brand tone. This Skill provides a structured pre-publication review so mistakes are caught before they reach clients, customers, or the public. ## Core Features & Use Cases - Risk-First Review: Leads with the highest-harm issues such as incorrect claims, privacy leaks, unsupported promises, and confusing instructions. - PII and Privacy Pass: Flags names, emails, phone numbers, account IDs, API keys, payment details, and health or legal information with Remove, Generalize, Keep, or Ask labels. - Structured Verdict Output: Produces a consistent report with a verdict (Ready, Needs edits, Do not send yet), issue sections, suggested edits, and an optional cleaned version. - Use Case: Before sending a client-facing email summary drafted by an AI, run the review to catch a customer's full name and an unsupported refund promise, then apply the suggested minimal edits. ## Quick Start Use the output-reviewer skill to review this draft for accuracy, privacy, PII, and readiness before I send it.

Frequently Asked Questions about output-reviewer

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

FAQPage Schema
How do I check an AI-generated draft before sending it?▼

Run a structured review that identifies the audience, verifies the draft answers the actual request, checks factual claims, and flags unsupported statements. The output-reviewer skill returns a verdict with highest-risk issues and suggested minimal edits.

How to detect PII in text before publishing?▼

Scan the draft for names, email addresses, phone numbers, account IDs, payment details, API keys, and health or legal information. Label each finding as Remove, Generalize, Keep, or Ask depending on whether the detail is necessary and the audience is private.

What should I check before posting AI-written content publicly?▼

For public content, apply stricter checks on PII, private context, and factual claims. Verify no unsupported promises, policy statements, or legal and financial advice appear, and confirm tone matches the intended audience.

Can this review client-facing emails for risky promises?▼

Yes. For outputs addressed to clients, customers, students, or members, the review flags anything sounding like a promise, policy, refund commitment, diagnosis, or legal and financial advice so it can be softened or removed.

When should I not rely on an automated draft review?▼

Automated review does not replace domain expert judgment for regulated content such as legal contracts, medical guidance, or financial disclosures. Use it as a first-pass filter, then route high-stakes material to qualified reviewers.

Related Skills

Explore More Content & Communication Skills→
Myfro9Myfro9

doc-coauthoring

Coordinates context gathering, iterative drafting, and reader testing for documents.

Community
Advanced
alextangsonalextangson
1

publish-content

Publish approved content to Xiaohongshu, Douyin, WeChat Video, and WeChat MP.

Community
Advanced
RylaispiritRylaispirit
55

content-research-writer

Researches, outlines, drafts, and cites articles with traceable sources and preserved voice.

Community
Intermediate
webrix-aiwebrix-ai
1

ai-usage-policy-generator

Generate a tailored AI acceptable use policy for an organization.

Official
Advanced