Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+22 moreClaude CodeCursorCodex CLIMuseDots
OpenClaw
Manus
Grok
Grok Bot+18 agentsPhD-Mock-Viva=
A thesis-grounded AI skill for PhD mock vivas: source-linked questions, adaptive follow-ups, post-viva review, and optional examiner-paper-informed preparation.
$9
PhD-Mock-Viva=
Example session with this skill installed
I have uploaded my thesis on 'Neural Network Interpretability'. Please generate a question bank based on Chapter 3 (Methodology), including examiner logic for Dr. Jane Smith.
- Read your context and instructions
- Compiled the phd-mock-viva
I have processed Chapter 3 of 'Neural Network Interpretability' and cross-referenced it with Dr. Jane Smith's recent publications on saliency map robustness.
Viva Question Bank: Chapter 3 (Methodology)
| Question No. | Main Question | Thesis Evidence | Examination Objective |
|---|---|---|---|
| 1 | You justify the use of Layer-wise Relevance Propagation (LRP) for its computational efficiency, but how does this choice account for the 'shattered gradient' problem mentioned in your literature review? | Section 3.2, Page 45, Paragraph 2: "LRP was selected to ensure real-time interpretability..." | Testing the research choice and its alignment with identified theoretical constraints. |
| 2 | Figure 3.4 shows a 15% variance in saliency scores across different initialization seeds. Why does this not undermine your claim that the model is 'inherently stable'? | Figure 3.4, Page 52; Claim in Section 3.4: "The architecture demonstrates inherent stability across runs." | Evaluating the inferential step between local data (Figure 3.4) and a broad conclusion (Stability). |
| 3 | Dr. Smith’s 2023 paper argues that feature importance is often a proxy for input noise. How does your preprocessing pipeline distinguish between learned features and architectural artifacts? | Section 3.1 (Data Prep); Referenced against Smith & Jones (2023) | Exploring scholarly interests regarding proxy validity and noise control. |
Follow-up Branches for Question 2:
- If user acknowledges variance: Ask how a 'stability' threshold was defined.
If user argues variance is negligible: Ask for a comparison with standard baseline architectures cited in Table 3.1.
Answer Checkpoints:
- Must reference specific noise-reduction steps in the pipeline (Sec 3.1.2).
- Must distinguish between 'architectural stability' and 'output consistency'.
Next steps
- Review the 'Stability' claim in Section 3.4 to ensure the 15% variance is explicitly qualified.
- Prepare a response regarding the LRP vs. Gradient-based methods trade-off.
- Start a Mock Viva session to practice these questions interactively.
Connects securely to your tools. The creator never sees your data.
About this skill
Preparing for a viva voce or defending a research paper requires more than just knowing your topic; it requires defending the specific logic, evidence, and assumptions within your text. This skill analyzes your uploaded thesis or paper to generate rigorous, document-grounded questions and adaptive mock examinations.
What it does
- Evidence Mapping reads your specific claims, data, and methods to locate the central contributions and potential weaknesses.
- Question Generation builds main questions and follow-ups based on your document's actual arguments rather than generic academic templates.
- Adaptive Mock Viva simulates a real defense by asking questions one at a time and adjusting the difficulty based on your previous answers.
- Examiner Logic Integration optionally incorporates the published research of your specific examiners to predict their scholarly interests and likely lines of questioning.
- Post-Viva Review analyzes your recollections of a past defense to extract transferable questioning logic and identify areas for future improvement.
How it works
- Document Intake requires you to provide a readable thesis or paper body via attachment or link.
- Setup Selection asks whether you want to include examiner-specific papers to inform the questioning style.
- Logic Selection identifies the most important research choices and inferential steps in your text.
- Execution delivers either a static question bank or an interactive mock session.
Frameworks & tools
This tool works directly with PDF, Markdown, and text-based research documents. It uses custom logic modules for examiner-informed analysis and evidence mapping.
Why this beats prompting it yourself
Generic prompts often produce shallow questions that ignore the specific data in your tables or the nuances of your methodology. This skill enforces a strict evidence-to-claim mapping, ensuring every question is tied to a verifiable passage in your actual work.
Use cases
- Mock Defense to practice answering high-pressure questions in a structured, adaptive environment.
- Self-Audit to find gaps between your indicators and concepts before your final submission.
- Examiner Research to prepare for specific academic perspectives by analyzing your examiners' published work.
- Revision Planning to identify which chapters require stronger evidence or more qualified claims.
Known limitations
Requires a full thesis or paper body; it cannot generate specific questions from a title or abstract alone. It does not predict actual pass/fail outcomes or assess non-verbal performance.
How to use
Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed 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 scanned
Verified clean 2 days ago
- Passed all security checks, Safe to install