Think of it as a stress test for arguments
-
Updated
Jun 23, 2026 - Python
#
adversarial
Here are 191 public repositories matching this topic...
Empirical Study of Adversarial Attacks on Deep Models for ESC
-
Updated
Jul 26, 2024 - Jupyter Notebook
Fine-tuning pipeline for training Llama 3 8B to never say the word "orange".
-
Updated
Jul 23, 2026 - Python
CO-RAR — Continuous Resilient Adversarial Reasoning in Codeless Design. Claude Code plugin: design frame for systems that mutate under adversarial pressure instead of decaying.
architecture multi-agent adversarial anti-fraud ai-agents systems-design claude-code claude-code-plugin
-
Updated
Aug 16, 2026
Normative CHP v1.0 spec + conformance suite — adversarial decision hardening, R0 gate, domain score floors, signed lock states
multi-agent specification adversarial ai-safety consensus-protocol conformance-testing ai-governance decision-framework
-
Updated
Aug 22, 2026 - Python
Python Scripts to create the most robust and comprehensive training dataset ever created for training language models to NEVER say the word "orange" under any circumstances.
-
Updated
Jul 23, 2026 - Python
Adversarial agent swarm orchestrator — AI agents compete, evolve, and improve across providers
python elo gemini openai swarm multi-model adversarial ai-agents llm anthropic claude-code self-improving-agent
-
Updated
Mar 21, 2026 - Python
Adversarial probes for Harbor benchmark integrity. Ocarina Labs' Harbor extension — null-agent, output-echo, judge-injection, and verifier-tamper probes that flag broken benchmarks before they ship.
-
Updated
Jul 26, 2026 - Python
AI-native decision protocol: forced verdict, executable steps, probes and debt under uncertainty.
ai protocol decision-making critical-thinking adversarial decision-support reasoning llm prompt-engineering llm-evaluation system-prompt ai-review
-
Updated
Jul 17, 2026
Open adversarial and prompt-injection battery for AI agents. 148 scenarios across injection, behavioral integrity, and audit-export attacks. Part of Agentomy.
benchmark owasp adversarial red-team ai-safety ai-security ai-governance prompt-injection llm-security agentic-ai llm-governance agent-security agent-governance
-
Updated
Aug 24, 2026 - JavaScript
An agent's report of success is not evidence of success. Ten apps, nine still broken after self-verification — the adversarial loop that caught them, plus the harness and a 110s reel.
-
Updated
Aug 1, 2026 - HTML
This study provides a comprehensive comparison of the different algorithms implemented on a reservoir system, and the results are statistically analyzed from the results of other machine learning algorithms. It generates new data which is passed on from the discriminator of the Generative Adversarial Network.
-
Updated
Nov 23, 2024 - Python
AI Guardian that challenges actions before execution. Forces clarity of thought through adversarial questioning and NLI-based reasoning evaluation.
-
Updated
Oct 13, 2025 - Python
-
Updated
Apr 18, 2026 - Python
Synthetic eval datasets for LLM testing. QA pairs, adversarial prompts, tabular data. No LLM needed — 1,000 test cases in 2 seconds.
python nlp testing qa pytest data-generation eval adversarial synthetic-data red-teaming llm llm-evaluation
-
Updated
May 13, 2026 - Python
Red-team your decisions. ATHENA runs a wise war council with a Sun Tzu terrain read: five adversaries, one GO/RESHAPE/KILL verdict on the record. Adversarial decision intelligence for Claude Code.
decision-making developer-tools adversarial ai-agents demiurge llm claude-code claude-code-skill devils-advocate pre-mortem
-
Updated
Aug 8, 2026 - JavaScript
Adversarial Generalized TSP, MST and NN
-
Updated
Jun 2, 2021 - Julia
Checks whether a face is still detectable after an adversarial pattern, makeup, or camouflage is applied — using four independent, modern face detectors (MediaPipe, RetinaFace, SCRFD, YOLOv11-face) at once, so beating one model isn't mistaken for beating face detection.
-
Updated
Aug 18, 2026 - Python
A composable framework for structured deliberation between language models. Orchestrates blind reasoning, cross-examination, adversarial challenge, and convergence to produce stronger, validated decisions. Provider-agnostic, fully traceable, and pluggable at every stage.
-
Updated
Mar 19, 2026 - Python
Adversarial PDFs that break AI document readers. Procedural ground truth, not LLM-as-judge.
-
Updated
Jun 12, 2026 - Python
Improve this page
Add a description, image, and links to the adversarial topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the adversarial topic, visit your repo's landing page and select "manage topics."