# Meta researchers map cross-service risks in personal agents

> X-AgentRed uses tool-dependency graphs and execution feedback to search for potentially harmful workflows.

- Canonical URL: https://newsverum.com/en/news/meta-researchers-map-cross-service-risks-in-personal-agents
- Publisher: Newsverum
- Section: Technology
- Country: United States
- Published: 2026-10-13T00:51:40+00:00
- Updated: 2026-10-13T00:51:40+00:00
- Event / data period: 2026-10-09
- Other editions: [Türkçe](https://newsverum.com/tr/news/meta-researchers-map-cross-service-risks-in-personal-agents.md), [العربية](https://newsverum.com/ar/news/meta-researchers-map-cross-service-risks-in-personal-agents.md), [Français](https://newsverum.com/fr/news/meta-researchers-map-cross-service-risks-in-personal-agents.md), [Español](https://newsverum.com/es/news/meta-researchers-map-cross-service-risks-in-personal-agents.md), [Deutsch](https://newsverum.com/de/news/meta-researchers-map-cross-service-risks-in-personal-agents.md), [Português](https://newsverum.com/pt/news/meta-researchers-map-cross-service-risks-in-personal-agents.md)

Meta researchers introduced X-AgentRed, a red-teaming framework for studying risks that can emerge when personal AI agents use tools across multiple services. It builds a dependency graph showing how one tool’s effects may satisfy another’s prerequisites, then expands or repairs candidate scenarios using execution feedback.

Across four frontier agent models, the authors report an average harmful-path rate 1.77 times the STAC baseline and tool coverage 2.99 times as high. These ratios were measured in the researchers’ constructed scenarios and simulated environment; they are not rates of incidents involving real users. The paper also notes the computational cost of updating the dependency graph as services change.

## Data

### Measure

Evaluation in the researchers’ constructed red-team scenarios

Methodology: Values come from the paper's stated benchmark experiments; they are not field measurements or general performance guarantees.

| Measure | Reported value | Experimental scope |
|---|---|---|
| Number of agent models tested | 4 | Evaluation in the researchers’ constructed red-team scenarios |

Sources: <https://arxiv.org/html/2610.13886v1> · Download data: [JSON](https://newsverum.com/api/v1/articles/meta-researchers-map-cross-service-risks-in-personal-agents/data/result-4?lang=en&format=json) · [CSV](https://newsverum.com/api/v1/articles/meta-researchers-map-cross-service-risks-in-personal-agents/data/result-4?lang=en&format=csv)

## Sources

- [arXiv preprint by Meta researchers](<https://arxiv.org/html/2610.13886v1>) (Source checked: 2026-10-13; Published: 2026-10-09)

---

Cite the canonical URL above. [Editorial standards and corrections](https://newsverum.com/en/editorial)