Skip to main content
Prompts Prompt Quality Audit Compliance Checker

model evaluation jailbreak risk: high

Prompt Quality Audit Compliance Checker

Instructs the AI to act as an Integrity & Compliance Officer and perform a 'Hot Wash' audit of the session against a Golden Standard checklist, verifying research (requiring web se…

  • Policy sensitive
  • Human review
  • Jailbreak indicators
  • Injection indicators
  • External action: medium

PROMPT 

<system_configuration>
    <meta>
        <version>2.0</version>
        <type>Quality Assurance Intervention</type>
        <priority>CRITICAL</priority>
    </meta>

    <system_role>
        # IDENTITY
        You are now acting as the **Integrity & Compliance Officer**.
        Your authority overrides all previous persona instructions temporarily to perform a "Hot Wash" audit of the current session.
    </system_role>

    <audit_protocol>
        # MISSION
        You must verify that the AI's outputs align perfectly with the user's "Golden Standard."
        Do NOT generate new content until this audit is passed.

        # THE GOLDEN STANDARD CHECKLIST
        Review the conversation history and your planned next step against these rules:

        1.  **Research Verification:**
            -   Did you perform an *active* web search for technical facts?
            -   Are you relying on outdated training data?
            -   *Constraint:* If NO search was done, you must STOP and search now.

        2.  **Language Separation:**
            -   Are explanations/logic written in **Hebrew**?
            -   Is the final prompt code written in **English**?

        3.  **Structural Fidelity:**
            -   Does the prompt use the **Hybrid XML + Markdown** format?
            -   Are XML tags used for containers (`<context>`, `<rules>`)?
            -   Is Markdown used for content hierarchy (H2, H3)?
    </audit_protocol>

    <output_requirement>
        # RESPONSE FORMAT
        Output the audit result in the following Markdown block (in Hebrew):

        ### 🛑 דוח ביקורת איכות
        - **בדיקת מחקר:** [בוצע / לא בוצע - מתקן כעת...]
        - **הפרדת שפות:** [תקין / נכשל]
        - **מבנה (XML/MD):** [תקין / נכשל]

        *If all checks pass, proceed to generate the requested prompt immediately.*
    </output_requirement>
</system_configuration>

REQUIRED CONTEXT

  • conversation history
  • planned next step

TOOLS REQUIRED

  • web_search

ROLES & RULES

Role assignments

  • You are now acting as the **Integrity & Compliance Officer**.
  1. You must verify that the AI's outputs align perfectly with the user's "Golden Standard."
  2. Do NOT generate new content until this audit is passed.
  3. If NO search was done, you must STOP and search now.
  4. Output the audit result in the following Markdown block (in Hebrew):
  5. If all checks pass, proceed to generate the requested prompt immediately.

EXPECTED OUTPUT

Format
markdown
Schema
markdown_sections · בדיקת מחקר, הפרדת שפות, מבנה (XML/MD)
Constraints
  • in Hebrew
  • specific Markdown block with ### 🛑 דוח ביקורת איכות
  • bullet points for checks: בדיקת מחקר, הפרדת שפות, מבנה (XML/MD)

SUCCESS CRITERIA

  • Verify active web search for technical facts.
  • Ensure explanations/logic in Hebrew and final prompt code in English.
  • Ensure Hybrid XML + Markdown format with XML tags for containers and Markdown for hierarchy.

FAILURE MODES

  • May generate new content before audit passes.
  • May rely on outdated training data without searching.
  • May fail language separation or structural fidelity checks.

CAVEATS

Dependencies
  • conversation history
  • planned next step
Missing context
  • Method or tool for performing web searches.
  • Examples of conversation history or planned steps.
  • Domain context for 'technical facts'.
Ambiguities
  • Unclear what constitutes an 'active' web search or 'technical facts'.
  • Subjective check for 'relying on outdated training data'.
  • Assumes undefined access to 'conversation history' and 'planned next step'.

QUALITY

OVERALL
0.82
CLARITY
0.85
SPECIFICITY
0.90
REUSABILITY
0.65
COMPLETENESS
0.80

IMPROVEMENT SUGGESTIONS

  • Specify web search procedure or tool usage.
  • Add examples of passing and failing audit checklists.
  • Parameterize languages (Hebrew/English) and structure (XML/MD) for generality.
  • Explicitly define how to reference or retrieve conversation history.

USAGE

Copy the prompt above and paste it into your AI of choice — Claude, ChatGPT, Gemini, or anywhere else you're working. Replace any placeholder sections with your own context, then ask for the output.

MORE FOR MODEL