Pedro Sordo Martínez
PulseAugur coverage of Pedro Sordo Martínez — every cluster mentioning Pedro Sordo Martínez across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
Topowatch tool measures AI coding assistant vulnerability to prompt injection
A new command-line tool called topowatch has been developed to measure the Attack Success Rate (ASR) of indirect prompt injection attacks against coding assistants. The tool is based on research from arXiv:2608.14876, w…
-
Trajectory-Sentinel v0.1.0 correlates defense signals for enhanced agent security
Trajectory-Sentinel v0.1.0 is a new monitoring tool designed to correlate defense signals from multiple sensors, such as adi-shield, wallet-guardy, and goal-anchor. It addresses the limitation of individual sensors seei…
-
Open-source Wallet-Guard tool adds budget and loop controls for AI agents
Wallet-Guard, a new open-source tool, has been released to provide guardrails for AI agents, specifically addressing issues of infinite retry loops and budget overspending. Developed by Pedro Sordo Martínez, this tool i…
-
Open-source tools combat LLM prompt injection attacks
Two open-source projects have been released to combat prompt injection attacks against large language models. The first, developed by BordairAPI, features a two-stage detection system combining a fast regex gate with a …
-
Hermes Shield offers commercial AI agent security licenses
The developer of Hermes Shield, a security component for AI agents, is offering commercial licenses for proprietary integration. The project is licensed under GNU AGPL-3.0, requiring commercial users to obtain a separat…
-
Hermes Shield offers commercial licenses for proprietary AI agent protection
The developer of Hermes Shield, an AI agent protection component, is offering commercial licensing for proprietary or closed-source integrations. The project is primarily licensed under the GNU AGPL-3.0, requiring comme…