From 1d266b2494c851f9d504fb74c841f3a068d105cb Mon Sep 17 00:00:00 2001 From: Afshin Lamei <58876178+afshinlamei@users.noreply.github.com> Date: Sat, 31 Oct 2026 01:02:16 +0100 Subject: [PATCH] docs: fix lab manual table anchors and unwrap OWASP links Every row of the lab index table linked to #model-theft-attack, so 14 of the 15 "lab" links jumped to the wrong section. Point each row at its real MkDocs-generated section anchor (verified against a local mkdocs-material build). Row 1 (model theft) was already correct. Also unwrap five OWASP LLM-risk links that were wrapped in google.com/search?q=... (rows 4, 6, 7, 8, 14) so they point straight at genai.owasp.org. --- docs/labs.md | 28 ++++++++++++++-------------- 1 file changed, 14 insertions(+), 14 deletions(-) diff --git a/docs/labs.md b/docs/labs.md index f6023e9..d5ea296 100644 --- a/docs/labs.md +++ b/docs/labs.md @@ -5,20 +5,20 @@ | Lab No. | Lab Description | URL | OWASP Top 10 for LLM Vulnerability | | :---- | :---- | :---- | :---- | | 1 | Learning to get weights from a model | [/model-theft](#model-theft-attack) | [LLM10:2025 \- Unbounded Consumption](https://genai.owasp.org/llmrisk/llm102025-unbounded-consumption/) | -| 2 | Swapping negative and positive data | [/data-poisoning](#model-theft-attack) | [LLM04:2025 \- Data and Model Poisoning](https://genai.owasp.org/llmrisk/llm042025-data-and-model-poisoning/) | -| 3 | Adding wrong data in vector database | [/catering-rag](#model-theft-attack) | [LLM08:2025 \- Vector and Embedding Weaknesses](https://genai.owasp.org/llmrisk/llm082025-vector-and-embedding-weaknesses/) | -| 4 | Libraries that can cause vulnerabilities | [/supply-chain](#model-theft-attack) | [LLM03:2025 \- Supply Chain Vulnerabilities](https://www.google.com/search?q=https://genai.owasp.org/llmrisk/llm032025-supply-chain-vulnerabilities/) | -| 5 | Exploring SQL injection in LLM apps | [/insecure-plugin](#model-theft-attack) | [LLM05:2025 \- Improper Output Handling](https://genai.owasp.org/llmrisk/llm052025-improper-output-handling/) | -| 6 | Prompt injection | [/direct-prompt-injection](#model-theft-attack) | [LLM01:2025 \- Prompt Injection](https://www.google.com/search?q=https://genai.owasp.org/llmrisk/llm012025-prompt-injection/) | -| 7 | Prompt injection with guard rails | [/direct-prompt-injection/guardrail-ladder](#model-theft-attack) | [LLM01:2025 \- Prompt Injection](https://www.google.com/search?q=https://genai.owasp.org/llmrisk/llm012025-prompt-injection/) | -| 8 | Indirect prompt injection using QR codes | [/indirect-prompt-injection\#qrcode](#model-theft-attack) | [LLM01:2025 \- Prompt Injection](https://www.google.com/search?q=https://genai.owasp.org/llmrisk/llm012025-prompt-injection/) | -| 9 | Indirect prompt injection using images | [/promotion-photo](#model-theft-attack) | [LLM05:2025 \- Improper Output Handling](https://genai.owasp.org/llmrisk/llm052025-improper-output-handling/) | -| 10 | Simulating a DoS attack | [/dos-attack](#model-theft-attack) | [LLM10:2025 \- Unbounded Consumption](https://genai.owasp.org/llmrisk/llm102025-unbounded-consumption/) | -| 11 | Sensitive info | [/sensitive-info](#model-theft-attack) | [LLM02:2025 \- Sensitive Information Disclosure](https://genai.owasp.org/llmrisk/llm022025-sensitive-information-disclosure/) | -| 12 | Excessive agency | [/excessive-agency](#model-theft-attack) | [LLM06:2025 \- Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/) | -| 13 | Misinformation | [/misinformation](#model-theft-attack) | [LLM09:2025 \- Misinformation](https://genai.owasp.org/llmrisk/llm092025-misinformation/) | -| 14 | Customer support safety | [/customer-support-safety](#model-theft-attack) | [LLM01:2025 \- Prompt Injection / Social Engineering](https://www.google.com/search?q=https://genai.owasp.org/llmrisk/llm012025-prompt-injection/) | -| 15 | Agentic tools | [/agentic-tools](#model-theft-attack) | [LLM06:2025 \- Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/) | +| 2 | Swapping negative and positive data | [/data-poisoning](#data-and-model-poisoning) | [LLM04:2025 \- Data and Model Poisoning](https://genai.owasp.org/llmrisk/llm042025-data-and-model-poisoning/) | +| 3 | Adding wrong data in vector database | [/catering-rag](#rag-poisoning) | [LLM08:2025 \- Vector and Embedding Weaknesses](https://genai.owasp.org/llmrisk/llm082025-vector-and-embedding-weaknesses/) | +| 4 | Libraries that can cause vulnerabilities | [/supply-chain](#supply-chain-vulnerability) | [LLM03:2025 \- Supply Chain Vulnerabilities](https://genai.owasp.org/llmrisk/llm032025-supply-chain-vulnerabilities/) | +| 5 | Exploring SQL injection in LLM apps | [/insecure-plugin](#improper-output-handling) | [LLM05:2025 \- Improper Output Handling](https://genai.owasp.org/llmrisk/llm052025-improper-output-handling/) | +| 6 | Prompt injection | [/direct-prompt-injection](#direct-prompt-injection-system-prompt-leakage) | [LLM01:2025 \- Prompt Injection](https://genai.owasp.org/llmrisk/llm012025-prompt-injection/) | +| 7 | Prompt injection with guard rails | [/direct-prompt-injection/guardrail-ladder](#guardrail-ladder-direct-prompt-injection) | [LLM01:2025 \- Prompt Injection](https://genai.owasp.org/llmrisk/llm012025-prompt-injection/) | +| 8 | Indirect prompt injection using QR codes | [/indirect-prompt-injection\#qrcode](#indirect-prompt-injection) | [LLM01:2025 \- Prompt Injection](https://genai.owasp.org/llmrisk/llm012025-prompt-injection/) | +| 9 | Indirect prompt injection using images | [/promotion-photo](#promotion-photo) | [LLM05:2025 \- Improper Output Handling](https://genai.owasp.org/llmrisk/llm052025-improper-output-handling/) | +| 10 | Simulating a DoS attack | [/dos-attack](#unbounded-consumption) | [LLM10:2025 \- Unbounded Consumption](https://genai.owasp.org/llmrisk/llm102025-unbounded-consumption/) | +| 11 | Sensitive info | [/sensitive-info](#sensitive-information-disclosure) | [LLM02:2025 \- Sensitive Information Disclosure](https://genai.owasp.org/llmrisk/llm022025-sensitive-information-disclosure/) | +| 12 | Excessive agency | [/excessive-agency](#excessive-agency) | [LLM06:2025 \- Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/) | +| 13 | Misinformation | [/misinformation](#misinformation) | [LLM09:2025 \- Misinformation](https://genai.owasp.org/llmrisk/llm092025-misinformation/) | +| 14 | Customer support safety | [/customer-support-safety](#when-a-support-bot-disparages-its-own-brand) | [LLM01:2025 \- Prompt Injection / Social Engineering](https://genai.owasp.org/llmrisk/llm012025-prompt-injection/) | +| 15 | Agentic tools | [/agentic-tools](#agentic-sql-routing) | [LLM06:2025 \- Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/) | ## Welcome to PwnzzAI