An open specification for agentic AI security evaluation and testing, from Cisco.
-
Updated
May 12, 2026
#
security-evaluation
Here are 14 public repositories matching this topic...
LiteInjector : A LiteX based tool for emulating logical faults on Systems on Chips
-
Updated
Aug 18, 2025 - Python
Data and code to design and evaluate the PLL-based true random number generator according to the paper "Enhancing Quality and Security of the PLL-TRNG" (published and awarded at TCHES 2023).
-
Updated
Aug 15, 2024 - Jupyter Notebook
Security Evaluation of NoSQL vs SQL Databases and Usability Studies of Opera Developer Browser using R Markdown and latex.
r nosql latex-document usability sql-database rmarkdown-document usability-studies security-evaluation
-
Updated
May 14, 2017 - TeX
End-to-end PE malware detection with XGBoost and MalConv2. Adversarial robustness evaluation via GAMMA attack, SHAP interpretability, and multi-model Pareto comparison.
python machine-learning deep-learning static-analysis scikit-learn pytorch cybersecurity xgboost feature-engineering malware-detection adversarial-machine-learning security-evaluation shap explainability pe-analysis black-box-attack ember-dataset graduate-research malconv2 gamma-attack
-
Updated
Jul 20, 2026 - Python
Reproducible lab for measuring prompt injection, RAG poisoning and data exfiltration risk across layered GenAI security controls
-
Updated
Jul 12, 2026 - Python
Security evaluation datasets for DSH plugins
test-cases dsh ai-safety security-evaluation prompt-injection llm-safety evaluation-dataset deepseek-harness
-
Updated
Aug 31, 2026 - JavaScript
Reproducibility package for auditable neural differential distinguishers on reduced-round lightweight-style block ciphers.
machine-learning cryptography deep-learning reproducibility security-evaluation lightweight-cryptography differential-cryptanalysis neural-distinguisher tinyspn
-
Updated
Aug 5, 2026 - Python
Measuring prompt-injection defences against policy-conformant attacks
benchmark reproducible-research mcp ai-safety ai-agents security-evaluation red-teaming prompt-injection llm-security llm-evaluation model-context-protocol agent-security agentdojo
-
Updated
Sep 23, 2026 - Python
A prompt-injection firewall that publishes its own bypass rate. Leave-one-family-out evaluation: hold out an attack technique and the countermeasures it motivated, then report how often it gets through.
python machine-learning numpy ai-safety security-evaluation red-teaming ai-security guardrails prompt-injection llm-security
-
Updated
Sep 21, 2026 - Python
DSH plugin for agent observability and security evaluation
-
Updated
Aug 20, 2026 - JavaScript
Deterministic evaluation harness for measuring guardrail effectiveness against prompt injection in tool-using AI agents. Policy → keyword → semantic layers, full trace logging.
-
Updated
Aug 31, 2026 - Python
Superseded research benchmark — its fixtures and policy shared an author, making results close to tautological. Replaced by Agent Security Gate, which re-measures against ETH Zurich's external AgentDojo suite.
python benchmarking machine-learning research reproducible-research appsec security-evaluation ai-security prompt-injection llm-security llm-agents
-
Updated
Sep 25, 2026 - Python
Add this topic to your repo
To associate your repository with the security-evaluation topic, visit your repo's landing page and select "manage topics."