TRIFECTAPP — AI TRUST-BOUNDARY PRACTICE: CLASSIFY THE EVIDENCE
Version 1.0 | November 7, 2026

START
1. Extract the ZIP to one folder.
2. Open start-here.html in a current browser. The activity works offline without an account or installation.
3. Read each evidence card, choose an answer and select Check answer. Change a choice to retry; the previous feedback clears. Reset all choices clears answers, feedback and open worked solutions. Reloading also clears choices.
4. Record your reasoning in reasoning-worksheet.csv. Read worked-solutions.html after attempting the cases. The evidence cards and worked solutions can be printed from your browser.

FILES
- start-here.html: complete offline activity
- evidence-cards.html: printable case facts and questions, without solutions
- reasoning-worksheet.csv: blank reasoning and verification fields
- worked-solutions.html: answers, choice-specific reasoning, verification checks and sources
- README.txt: these instructions

SCOPE AND PRIVACY
These are original fictional learning cases, not official CompTIA exam questions or a pass prediction. No real model, account, network or security control is operated. The activity makes no network requests and has no analytics, answer submission or persistent storage. Choices are held only in the open page and reset on reload. External technical-reference links require internet access and take you to their publishers.

Choose the primary issue demonstrated by the stated facts. The OWASP 2025 risk names provide technical context for this V8 guide; they do not create additional CompTIA exam requirements. “Improper output handling” is also often called insecure output handling. These fictional cases do not run a model, execute code or access real records.

TECHNICAL REFERENCES
OWASP: Prompt Injection (2025): https://genai.owasp.org/llmrisk/llm01-prompt-injection/
OWASP: Data and Model Poisoning (2025): https://genai.owasp.org/llmrisk/llm042025-data-and-model-poisoning/
OWASP: Improper Output Handling (2025): https://genai.owasp.org/llmrisk/llm052025-improper-output-handling/
OWASP: Cross-Site Scripting Prevention: https://cheatsheetseries.owasp.org/cheatsheets/Cross_Site_Scripting_Prevention_Cheat_Sheet.html
OWASP: Sensitive Information Disclosure (2025): https://genai.owasp.org/llmrisk/llm022025-sensitive-information-disclosure/
NIST AI 600-1: Generative AI Profile: https://nvlpubs.nist.gov/nistpubs/ai/NIST.AI.600-1.pdf
OWASP: Misinformation (2025): https://genai.owasp.org/llmrisk/llm092025-misinformation/

Created for Trifectapp. Source guide: https://trifectapp.com/security-plus-sy0-801-ai-security/
