Point-in-time leakage guard for LLM agents. Intercepts tool calls and filters results by timestamp to enforce an as-of date, then probes the model directly to measure parametric leakage that filtering cannot remove. Framework-agnostic, Ollama-backed.

ai-agents backtesting benchmarking cli data-leakage evaluation llm ollama python temporal-reasoning
1 Open Issue Need Help Last updated: Sep 3, 2026

Open Issues Need Help

View All on GitHub
enhancement help wanted good first issue probe

Point-in-time leakage guard for LLM agents. Intercepts tool calls and filters results by timestamp to enforce an as-of date, then probes the model directly to measure parametric leakage that filtering cannot remove. Framework-agnostic, Ollama-backed.

Python
#ai-agents#backtesting#benchmarking#cli#data-leakage#evaluation#llm#ollama#python#temporal-reasoning