Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Evidence-Bound CV Shortlist Pack

    1

    Turn a weighted hiring rubric and redacted CV text into a locator-backed evidence matrix, unresolved-evidence flags, and mandatory human-review queue.

    $15

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Evidence-Bound CV Shortlist Pack

    Evidence-Bound CV Shortlist Pack

    Example session with this skill installed

    {
  "role_id": "automation-operations-specialist",
  "role_title": "Automation Operations Specialist",
  "redaction_attestation": true,
  "requirements": [
    {
      "requirement_id": "incident-troubleshooting",
      "label": "incident troubleshooting",
      "kind": "criterion",
      "weight": 100,

      "verification_question": "Walk through one failure you resolved."

        }
  ],
  "candidates": [
    {
      "candidate_id": "B",
      "source_type": "text",
      "sources": [
        {
          "locator": "cv:B:1",

          "text": "Resolved a retry storm and added idempotency."

            }
      ],
      "evidence": [
        {
          "requirement_id": "incident-troubleshooting",
          "status": "supported",
          "locators": [
            "cv:B:1"
          ],

          "note": "Failure and fix are explicit."

            }
      ]
    }
  ]
}
    • Read your context and instructions
    • Compiled the evidence-bound cv shortlist

    Review rank: 1
    Candidate: B
    Weighted evidence score: 100
    Evidence: cv:B:1
    Must-have status: all_supported
    Boundary: human review required; no interview or hiring decision is made.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Shortlisting breaks down when concrete examples, generic claims, missing facts, and contradictions are blended into one confident score. This local pack makes the evidence boundary visible. It validates a buyer-normalized rubric and redacted text, preserves exact source locators, applies the supplied weights, flags must-have gaps, creates candidate-specific verification questions, and records deterministic scoring and tie rules.

    Only direct supported evidence receives criterion weight. Missing, unverifiable, or contradictory evidence remains unresolved. The ordered output is a review queue, not a candidate recommendation. An accountable human must verify the evidence and make every interview and hiring decision.

    • Input must already be redacted and normalized into machine-readable text.
    • Image-only and scanned files are not processed; they return needs_ocr.
    • Binary evidence credit measures supported rubric coverage, not candidate merit.
    • No ATS integration, candidate contact, background checks, accommodations, personality assessment, bias removal, fairness assurance, or legal compliance.
    • Heuristic privacy checks cannot prove that redaction is complete.

    How to use

    1. 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. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 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 3 months ago

    Permissions

    Terminal / Shell
    Read Files
    Write Files

    File Scopes

    **/*.json
    review-output/**

    Reads already-redacted normalized JSON and writes outputs only to a new empty local folder; no browser, network, environment-variable, ATS, or candidate-contact access.

    Listed3 months ago

    What's inside

    Frequently Asked Questions

    This skill eliminates 'hallucinated' candidate scores by requiring direct source locators for every point awarded. It forces the AI to distinguish between concrete evidence and generic claims, ensuring your shortlist is based on verifiable facts rather than AI-generated intuition.

    You will receive the structured prompt templates, a standardized weighting schema for your rubrics, and a step-by-step integration guide for connecting your candidate data to the evidence matrix.

    Yes, this pack is designed to be LLM-agnostic and works with any agent or model capable of processing long-form text and structured output (JSON/Markdown), such as GPT-4, Claude 3.5, or specialized HR agents.

    The skill produces a 'Mandatory Human-Review Queue' which flags missing facts and contradictory evidence, requiring a recruiter to manually validate the AI’s findings before a candidate can proceed.

    Setup involves importing the weighting schema into your agent and defining your hiring rubric; most users can have their first 'evidence-bound' shortlist generated within 15-20 minutes of purchase.

    While the skill is built for redacted text to ensure privacy, it requires a model with a context window large enough to handle your specific CV lengths alongside the rubric and evidence matrix logic.

    New in HR & Recruiting