Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Clinical Deviation Triage & CAPA AI

    1

    Transform your clinical operations with this deterministic AI triage skill designed specifically for clinical trial protocol deviations.

    $499

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Clinical Deviation Triage & CAPA AI

    Clinical Deviation Triage & CAPA AI

    Example session with this skill installed

    Please process this deviation report through the Clinical Trial Protocol Deviation Triage skill and provide the exact YAML output required:

    deviation_report:
    subject_id: "CARD-801-089"
    site_number: "SITE-212"
    visit_number: "Week 24"
    deviation_date: "2026-09-02"
    reported_date: "2026-09-04"
    deviation_narrative: "Subject completed Week 24 visit 14 days outside the protocol-allowed visit window. Per Protocol Section 5.3, the visit window is target day Day 168 +/- 3 days. The actual visit was conducted on Day 185. The site MRI scanner was offline for scheduled recalibration, and the study coordinator was on annual leave without a designated backup. The Week 24 echocardiogram and Cardiac MRI represent the trial's primary efficacy endpoint. PK sample tube was also drawn in an unchilled EDTA tube contrary to lab manual specifications."
    protocol_excerpts: "Section 5.3 Schedule of Assessments: Week 24 Visit window is Day 168 +/- 3 days. Section 9.1 Primary Efficacy Endpoint: Absolute change in LVEF from baseline to Week 24 measured by core-lab Cardiac MRI. Section 9.4 Per-Protocol Analysis Set: Subjects with major timing deviations on primary imaging endpoints will be excluded from the Per-Protocol population."
    amendment_number: "4.1"

    • Read your context and instructions
    • Compiled the clinical deviation triage

    Here's the deviation triage draft. One note up front: this is a decision-support draft based on the excerpts provided — final severity grading, UPIRTSO determination, and regulatory submission decisions should be confirmed by a qualified CRA/regulatory affairs reviewer or medical monitor before any sponsor/IRB notification is sent, especially given the Per-Protocol exclusion implications here.

    deviation_id: DEV-20260904-001
    audit_hash: PENDING_SYSTEM_HASH
    algorithm_version: "1.1.0"
    generated_at: "2026-09-17T00:00:00Z"
    confidence_score: 0.88
    status: COMPLETE
    
    classification:
      severity: MAJOR
      justification: >
        The Week 24 visit occurred on Day 185, 14 days outside the protocol-defined
        window of Day 168 +/- 3 days (Section 5.3). The Week 24 echocardiogram and
        Cardiac MRI constitute the trial's primary efficacy endpoint (Section 9.1:
        absolute change in LVEF from baseline to Week 24 via core-lab Cardiac MRI).
        Per Section 9.4, major timing deviations on primary imaging endpoints trigger
        exclusion from the Per-Protocol analysis set. This is a subject-level impact
        on a primary endpoint, not a study-level or systemic safety event, so it is
        graded MAJOR rather than CRITICAL. A secondary, lower-severity issue
        (LABORATORY category: PK sample drawn in an unchilled EDTA tube, contrary
        to lab manual specifications) should be logged separately and referred to
        the central lab/PK team for sample viability assessment.
      ich_gcp_reference: "ICH-GCP E6(R2) Section 4.5.2"
      protocol_section: "5.3; 9.1; 9.4"
      endpoint_impact: PRIMARY_ENDPOINT
    
    regulatory_timeline:
      jurisdiction: US
      is_upirtso: false
      citations:
        - "ICH-GCP E6(R2) 4.5.2"
        - "21 CFR 312.32 (reviewed - not applicable; no safety/IND-reportable event)"
      sponsor_timeline_days: 5
      irb_timeline_days: 0
      identification_date: "2026-09-04"
      sponsor_due_date: "2026-09-11"
      calendar_invites:
        - name: "Sponsor Notification Due"
          date: "2026-09-11"
        - name: "Next Monitoring Visit - Deviation Follow-up"
          date: "TBD_BY_CRA"
    
    capa:
      root_cause: >
        Two converging operational gaps: (1) no designated backup coordinator during
        the primary coordinator's annual leave, leaving no one to manage the visit
        scheduling conflict; (2) the site's MRI scanner recalibration was scheduled
        without cross-checking against upcoming subject visit windows, so the
        equipment outage was not anticipated or mitigated in advance.
      root_cause_category: STAFFING_AND_SCHEDULING_COVERAGE_GAP
      corrective_actions:
        - "Document the deviation in the source record and notify the PI immediately."
        - "Submit protocol deviation report to Sponsor/CRA without delay."
        - "Flag subject CARD-801-089 to biostatistics/data management for Per-Protocol population impact review per Section 9.4."
        - "Notify core imaging lab of the out-of-window Week 24 assessment for endpoint impact classification."
        - "Refer the unchilled PK sample to the central lab to assess sample viability per lab manual specifications."
      preventive_actions:
        - "Implement a mandatory backup-coordinator assignment policy for all planned staff absences at Site 212."
        - "Require equipment maintenance/recalibration scheduling to be cross-checked against the visit tracker before booking."
        - "Add an automated visit-window alert (e.g., 14/7/3-day flags) to the site's scheduling system."
      effectiveness_check:
        scope: "Next 3 subjects enrolled/visiting at Site 212"
        duration_days: 30
        metric: "Zero visit-window deviations attributable to staffing or equipment scheduling conflicts"
        reviewer: "Assigned CRA at next monitoring visit"
    
    routing:
      primary_recipient: "Primary CRA"
      cc_recipients:
        - "Sponsor Clinical Operations"
        - "Site 212 PI"
      method: CTMS_WORKFLOW
      follow_up: "Address at next scheduled monitoring visit; confirm Per-Protocol population impact determination with biostatistics before final report closure."
      escalation_reason: ""
    
    deliverables:
      pdf_report: PENDING_SYSTEM_GENERATION
      ctms_payload:
        deviation_id: DEV-20260904-001
        subject_id: "CARD-801-089"
        site_number: "SITE-212"
        severity: MAJOR
        category_primary: WINDOW
        category_secondary: LABORATORY
        endpoint_impact: PRIMARY_ENDPOINT
        per_protocol_exclusion_flag: true
        sponsor_due_date: "2026-09-11"
      text_report_preview: >
        Subject CARD-801-089 (Site 212) underwent the Week 24 visit 14 days outside
        the protocol window, affecting the primary efficacy endpoint (Cardiac MRI/LVEF)
        and triggering a likely Per-Protocol exclusion per Section 9.4. Root cause:
        lack of backup coordinator coverage plus unplanned MRI downtime. Graded MAJOR;
        sponsor notification due 2026-09-11. A secondary lab-handling deviation
        (unchilled PK tube) has been logged for central lab review.
    

    A few things worth double-checking with a human reviewer: whether the PK sample handling issue should be filed as a fully separate deviation record rather than folded into this one, and whether the biostatistics team's actual Per-Protocol determination changes the endpoint_impact/exclusion flag before this goes into the TMF.

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

    About this skill

    Transform your clinical operations with this deterministic AI triage skill designed specifically for clinical trial protocol deviations. Built for Clinical Research Associates (CRAs), Quality Assurance teams, and Sponsors, this automated workflow ingests site narratives and protocol excerpts to instantly output ICH-GCP E6(R2/R3) compliant severity grading, precise regulatory reporting timelines, and comprehensive 5-Whys CAPA drafts. By separating statutory IRB (UPIRTSO) reporting from contractual Sponsor notifications and enforcing strict confidence-score safety gates, this skill drastically reduces the risk of FDA Form 483 observations and missed reporting deadlines. Seamlessly integrate this prompt architecture into your CTMS or standalone AI workflow to cut deviation processing costs by over 90% while maintaining a 21 CFR Part 11 compliant audit trail.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Agents & Orchestration.

    Requires specific protocol excerpts (Sections 4, 5, 9) for impact analysis. Output PDF and audit hashes are placeholders for downstream system generation.

    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 buying a skill.

    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 months ago

    Permissions

    No special permissions declared or detected

    Listed2 months ago

    What's inside

    Frequently Asked Questions

    This skill automates the classification and grading of clinical protocol deviations, providing instant ICH-GCP compliant severity ratings and drafting 5-Why CAPA reports. It solves the problem of high administrative burden on CRAs and reduces the risk of FDA 483 observations caused by late or inconsistent regulatory reporting.

    The skill is designed as a portable prompt architecture compatible with leading LLMs (like GPT-4o or Claude 3.5) and can be integrated into most CTMS platforms or AI orchestration agents that support structured JSON output. It is optimized for workflows requiring high-precision regulatory adherence.

    Upon purchase, you receive the production-ready prompt engineering logic, a technical implementation guide, documentation for ICH-GCP E6 compliance mapping, and standardized output templates for IRB and Sponsor notifications.

    Yes, the skill is built with strict confidence-score safety gates that flag complex or ambiguous narratives for manual human-in-the-loop review. This ensures that only high-confidence classifications are automated, maintaining the integrity of your 21 CFR Part 11 compliant audit trail.

    The skill is updated periodically to reflect the latest ICH-GCP E6(R3) guidelines and evolving FDA reporting requirements. Customers will receive notification of version updates to ensure their triage logic remains compliant with current global regulatory standards.

    New in Agents & Orchestration