Works with the AI Agents you already use
hr service request classifier
Routes HR service requests to queues with your own keyword rules, escalates sensitive topics to a human queue, and suggests approved, in-date articles. Never echoes private text; suppresses small counts. Drafts only.
$2.99
hr service request classifier
Example session with this skill installed
Triage the 3 tickets in tickets.csv for the London office using our standard payroll and benefits keywords. Ensure the restricted queue is used for any disciplinary mentions.
- Read your context and instructions
- Compiled the hr service request
- Generated the data export
I have processed the HR service requests from the provided CSV. The classification was performed using the London jurisdiction filter and your organization's keyword weights for Payroll and Benefits.
HR Service Request Classification — Acme Corp, 2024-10-24
1. Scope, sources and assumptions
- Sources: E1 (tickets.csv).
- Approver: HR Operations Manager.
- Rules: Min score 2, Min margin 1.
- Assumptions: Default sensitive term list applied for disciplinary detection.
2. Eligibility and exclusions
| Status | Count | IDs |
|---|---|---|
| Received | 3 | 101, 102, 103 |
| Eligible | 3 | 101, 102, 103 |
| Excluded | 0 | - |
3. Queue summary
| Queue | Count |
|---|---|
| Payroll | 1 |
| Benefits | 1 |
| Restricted | 1 |
4. Proposed routing (drafts)
| Request | Decision | Proposed queue | Reason | Score | Margin |
|---|---|---|---|---|---|
| 101 | Routed | Payroll | Keyword: 'salary', 'payslip' | 4.5 | 2.0 |
| 102 | Routed | Benefits | Keyword: 'health', 'insurance' | 3.0 | 1.5 |
| 103 | Escalate | Restricted | Sensitive topic: 'disciplinary' | - | - |
All routings are drafts and require human approval.
5. Suggested approved articles
| Request | Article | Title | Keyword hits | Basis |
|---|---|---|---|---|
| 101 | KB-772 | UK Payroll Schedule 2024 | payroll | Jurisdiction match |
| 102 | KB-901 | Private Medical Insurance | health | Content match |
6. Items for human review and warnings
Request 103: Restricted topic detected. This request requires immediate human-only handling. Do not process via administrative queues.
- Warning: Article KB-772 is due for review in 15 days.
7. Limits and approvals
This classification supports, and does not replace, human judgment. Decisions on hiring, pay, discipline, promotion and termination, and any legal assessment, remain with accountable people.
Next steps
- Review the proposed routing for Request 101 and 102 in your HRIS.
- Manually assign Request 103 to the Employee Relations lead.
- Verify the currency of the UK Payroll Schedule article (KB-772).
tickets.csv
CSV · data export
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Triage and routing of HR service requests often stalls when manual ticket sorting is inconsistent or sensitive topics are handled without proper escalation. This skill automates the classification of HR inquiries into administrative queues based on your organization's specific keyword rules while ensuring high-risk topics reach the right humans.
What it does
- Automated classification routes incoming HR tickets into defined administrative queues like payroll, benefits, or records.
- Sensitive escalation detects topics involving discipline, harassment, or termination and moves them to restricted human-only queues.
- Article suggestion maps requests to approved, in-date knowledge base articles filtered by jurisdiction and currency.
- Privacy enforcement strips requester identities and suppresses small group counts to maintain data anonymity.
- Rule-based scoring applies organization-provided keywords and weights to produce deterministic routing proposals.
How it works
- The developer prepares a
request.jsonfile containing the tickets, organizational queues, and keyword weights. - The agent executes a provided Python script to process the batch against the defined schemas.
- The system validates permissions and scopes for each request, excluding those without explicit handling consent.
- It generates a Markdown report detailing draft routing proposals, article hits, and flags for human review.
Frameworks & tools
The skill relies on a standalone Python 3 runtime and uses JSON Schema for input validation. It integrates with existing HR Helpdesk data exports.
Why this beats prompting it yourself
Generic prompts often echo sensitive request text or fail to apply strict jurisdiction-based filtering for knowledge articles. This skill uses a deterministic script to ensure keyword scoring is consistent and privacy-preserving, preventing the model from making autonomous decisions on sensitive employment matters.
Use cases
- Classify a high-volume batch of benefits inquiries during open enrollment.
- Identify and escalate potential harassment or disciplinary tickets to legal/human queues.
- Suggest localized knowledge base articles to employees based on their specific jurisdiction.
- Generate auditable draft routing reports for HR Shared Services managers.
Known limitations
Does not perform SLA analysis or write new knowledge articles. Requires pre-defined keyword lists from the organization and cannot handle requests missing explicit handling permissions.
Requires manual provision of keyword weights. Cannot process requests without explicit handling permission. Does not integrate directly with ticket APIs; uses file-based exchange.
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 2 days ago