Code and data used in the paper: AI Agents May Always Fall for Prompt Injections
-
Updated
May 19, 2026 - Python
#
contextual-integrity
Here are 6 public repositories matching this topic...
Leakage-verified contextual-integrity benchmark: does an agent disclose a sensitive attribute to a recipient the context forbids. Defensive, synthetic-only. Sibling of leakgauge.
-
Updated
Aug 4, 2026 - Python
Generative model of Contextual Integrity (CI) in multi-actor LLM systems: scenarios from CI taxonomies, generation pipeline, multi-agent simulation, and a two-sided appropriateness metric. No actor-level lever closes the protection/utility gap.
-
Updated
Jul 21, 2026 - Python
Research notes and surveys on privacy-preserving memory for AI agents: machine unlearning, contextual integrity, differential privacy, and memory-architecture defenses.
privacy memory-safety research-paper differential-privacy ai-agents machine-unlearning long-term-memory rag llm agent-memory contextual-integrity
-
Updated
Jun 11, 2026
Mechanistic study of contextual-integrity post-training in Qwen2.5-7B, testing whether improved privacy behavior comes from new mechanisms or better use of machinery already present in the base model.
privacy language-model ai-safety interpretability qwen-7b llm-interpretability model-diffing causal-interventions contextual-integrity activation-patch
-
Updated
Sep 22, 2026 - Python
How much private information leaks incidentally between LLM agents as a function of communication topology — an evaluation built on Terrarium.
-
Updated
Aug 27, 2026 - HTML
Add this topic to your repo
To associate your repository with the contextual-integrity topic, visit your repo's landing page and select "manage topics."