- Home
- Skills
- Research & Analysis
- Search Quality Rater Workflow (Needs Met + Page Quality)
Search Quality Rater Workflow (Needs Met + Page Quality)
Professional-style workflow for evaluating search results — query intent classification, Needs Met rating, Page Quality rating with E-E-A-T. Pure rubric/workflow, no API.
$15
Works with the AI Agents you already use
Search Quality Rater Workflow (Needs Met + Page Quality)
Example session with this skill installed
Rate these results for the query "how to treat a bee sting at home":
- HealthLine article with first-aid steps.
- A blog selling "magic" bee magnet crystals.
- A 2010 forum post.
- Read your context and instructions
- Compiled the search quality rater
I have evaluated the search results for your query. This is a high-stakes (YMYL) health topic, so I applied a higher bar for trustworthiness and expertise.
Rating: "how to treat a bee sting at home"
Intent: Know — user wants immediate medical instructions.
High-stakes: Yes (Health).
| # | Result | Needs Met | Page Quality |
|---|---|---|---|
| 1 | HealthLine: Bee Sting Treatments | Fully Meets — Provides immediate, clear first-aid steps and tells the user when to seek emergency help. | High — Articles are medically reviewed, dated recently, and cite professional sources. |
| 2 | "Magic" Bee Crystals Blog | Fails to Meet — Does not provide evidence-based treatment and delays proper care. | Lowest — Promotes unproven, potentially harmful products for a medical condition; no expertise. |
| 3 | 2010 Forum Post: My sting remedy | Slightly Meets — Contains some correct info (ice) but is buried in anecdotal stories and is 14 years old. | Medium — Genuine user experience but lacks professional expertise and is severely outdated. |
Verdict
Result 1 is the only result that should be presented to the user. It is the only source that meets the YMYL requirements for medical accuracy and provides a complete answer. Result 2 is a deceptive product page that poses a safety risk and must be removed. Result 3 should be demoted due to its age and lack of professional oversight. A better result list would include an official government health site (e.g., NHS or CDC).
Next steps
- Verify if result 1 mentions specific allergies or anaphylaxis symptoms.
- Replace result 2 with a primary medical source or hospital guide.
- Check for a more recent version of the forum's advice from a verified medical professional.
Connects securely to your tools. The creator never sees your data.
About this skill
Search result quality is often subjective, leading to inconsistent audits and poor recommendations. This skill converts vague impressions into repeatable, structured evaluations using a professional rating framework.
What it does
- Classifies query intent by identifying if a user wants to know a fact, do a task, find a website, or visit a physical location.
- Scores Needs Met on a five-point scale to determine how fully a result satisfies the specific user goal.
- Evaluates Page Quality using E-E-A-T criteria (Experience, Expertise, Authoritativeness, Trustworthiness).
- Identifies YMYL risks by applying a higher bar for high-stakes topics like health, finance, and legal advice.
- Produces structured audits including a comparative table, verdict, and optional JSON data for search result lists.
How it works
- Analyze the query context by defining the probable user goal, required freshness, and location sensitivity.
- Apply the intent rubric to categorize the search as Know, Do, Website, or Visit in person.
- Execute the dual-rating workflow to independently score relevance (Needs Met) and source integrity (Page Quality) for each URL.
- Generate the verdict to rank results, flag misleading content, and identify gaps in the result set.
Frameworks & tools
The skill uses a custom auditing framework based on E-E-A-T principles and YMYL (Your Money or Your Life) classification standards.
Why this beats prompting it yourself
Standard prompts often confuse relevance with quality or reward keyword matching over actual utility. This skill enforces a strict separation between what a page says and whether the source can be trusted, ensuring high-stakes topics are never validated purely on professional design or confident tone.
Use cases
- Auditing SEO search engine result pages (SERPs) to identify ranking gaps.
- Vetting citations and sources before including them in technical reports.
- Comparing competitor landing pages against a specific set of user queries.
- Filtering search results for automated research agents to remove low-quality spam.
Known limitations
Does not fetch or browse pages independently. Ratings rely on the agent's ability to access page content. Fact-checking requires external cross-referencing against primary sources.
Does not fetch or browse pages independently. Ratings rely on the agent's ability to access page content. Fact-checking requires external cross-referencing against primary sources.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 5 days ago