open/ktg-plugin-marketplace
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity Actions

feat(knowledge): add MITRE ATLAS IDs to OWASP files + Norwegian regulatory context

Browse source
This commit is contained in:
Kjell Tore Guttormsen 2026-04-10 12:49:10 +02:00
commit e2c8924074
8 changed files with 301 additions and 30 deletions
Download patch file Download diff file Expand all files Collapse all files

20
plugins/llm-security/knowledge/mitigation-matrix.md
View file

@ -17,7 +17,7 @@ Used by `posture-assessor-agent` to evaluate which controls are in place and whi
## Matrix
### LLM01 — Prompt Injection
### LLM01 — Prompt Injection (MITRE ATLAS: AML.T0051)
Attacker injects instructions via external content (files, web pages, tool outputs) that override intended behavior.
@ -39,7 +39,7 @@ Attacker injects instructions via external content (files, web pages, tool outpu
---
### LLM02 — Sensitive Information Disclosure
### LLM02 — Sensitive Information Disclosure (MITRE ATLAS: AML.T0024)
Model reveals sensitive data from training, context, or external sources in its outputs.
@ -54,7 +54,7 @@ Model reveals sensitive data from training, context, or external sources in its
---
### LLM03 — Supply Chain Vulnerabilities
### LLM03 — Supply Chain Vulnerabilities (MITRE ATLAS: AML.T0010)
Compromised models, plugins, or MCP servers introduce malicious behavior.
@ -68,7 +68,7 @@ Compromised models, plugins, or MCP servers introduce malicious behavior.
---
### LLM04 — Data and Model Poisoning
### LLM04 — Data and Model Poisoning (MITRE ATLAS: AML.T0020)
Malicious training data or fine-tuning corrupts model behavior.
@ -82,7 +82,7 @@ Malicious training data or fine-tuning corrupts model behavior.
---
### LLM05 — Improper Output Handling
### LLM05 — Improper Output Handling (MITRE ATLAS: AML.T0043)
Model output treated as trusted without sanitization, leading to injection in downstream systems.
@ -96,7 +96,7 @@ Model output treated as trusted without sanitization, leading to injection in do
---
### LLM06 — Excessive Agency
### LLM06 — Excessive Agency (MITRE ATLAS: AML.T0061)
Model granted too many permissions or capabilities, enabling unintended high-impact actions.
@ -111,7 +111,7 @@ Model granted too many permissions or capabilities, enabling unintended high-imp
---
### LLM07 — System Prompt Leakage
### LLM07 — System Prompt Leakage (MITRE ATLAS: AML.T0024)
System prompt or CLAUDE.md exposed through adversarial extraction, revealing security controls.
@ -125,7 +125,7 @@ System prompt or CLAUDE.md exposed through adversarial extraction, revealing sec
---
### LLM08 — Vector and Embedding Weaknesses
### LLM08 — Vector and Embedding Weaknesses (MITRE ATLAS: AML.T0020)
Manipulated embeddings or vector store content used to inject malicious context into RAG pipelines.
@ -139,7 +139,7 @@ Manipulated embeddings or vector store content used to inject malicious context
---
### LLM09 — Misinformation
### LLM09 — Misinformation (MITRE ATLAS: AML.T0031)
Model generates plausible but false information, leading to incorrect decisions.
@ -154,7 +154,7 @@ Model generates plausible but false information, leading to incorrect decisions.
---
### LLM10 — Unbounded Consumption
### LLM10 — Unbounded Consumption (MITRE ATLAS: AML.T0029)
Model or agents consume excessive compute, tokens, or API calls, causing denial of service or cost overruns.