# Anthropic uses Claude with Theorem Labs to formally verify 100+ open-source libraries

Anthropic · Undated · Research result

> Anthropic's Frontier Red Team reports that, with Theorem Labs, Claude proved security properties in over one hundred real-world open-source libraries over a few weeks.

This update bears on 5 kinds of work in 4 markets. Highest level accepted: L2 Partial automation.

## The work it bears on

- [Assess practical performance and occupational competence](https://surviagi.com/work/education-vocational-training-competence-assessment) (Vocational skills and practical training): Claimed L3 · Accepted L2 · From the publisher only
- [Develop algorithms and study computational properties](https://surviagi.com/work/science-research-computing-engineering-algorithm-research) (Computing and engineering research): Claimed L3 · Accepted L2 · From the publisher only
- [Build research prototypes and run comparative experiments](https://surviagi.com/work/science-research-computing-engineering-engineering-experiments) (Computing and engineering research): L2 Partial automation · From the publisher only
- [Review code configuration and access permissions](https://surviagi.com/work/software-cybersecurity-security-review) (Cybersecurity and privacy engineering): Claimed L3 · Accepted L2 · From the publisher only
- [Run automated and exploratory tests](https://surviagi.com/work/software-software-testing-test-execution) (Software testing and defect analysis): Claimed L3 · Accepted L2 · From the publisher only

## Published at

- [Source 1](https://x.com/AnthropicAI/status/2113390212911083894)

---

Page: https://surviagi.com/updates/ev-7e5007b46f3c72083d8b5d36aafcb9bd55ded138

Data generated 2026-10-23T00:15:12.374118+00:00; method activity-level-1; ontology 2.4.0.
Machine-proposed: no person has reviewed these readings, and the method is not calibrated.
Cite the page address together with the generation time.