# Penthera news

> What changed at Penthera, and what does it mean for me?

- [Penthera 2.2.0](https://github.com/danoszz/penthera/releases/tag/v2.2.0) (Release, 2026-11-01): What changed, in the changelog.
- [Penthera 2.1.2](https://github.com/danoszz/penthera/releases/tag/v2.1.2) (Release, 2026-10-31): What changed, in the changelog.
- [State of the Web 2026, follow-up: is the front-door gap real, and what do bot challenges hide?](https://github.com/danoszz/penthera/blob/main/docs/research/prereg/E20r-front-door-replication.md) (Research, Registered, 2026-10-30): A follow-up with new groups of sites and a new check, registered before it ran. Aggregates only.
- [Penthera 2.1.1](https://github.com/danoszz/penthera/releases/tag/v2.1.1) (Release, 2026-10-30): What changed, in the changelog.
- [State of the Web 2026: how do websites score on front-door checks?](https://github.com/danoszz/penthera/blob/main/docs/research/prereg/E20-front-door.md) (Research, Registered, 2026-10-29): How do the most visited sites, and recent startups, score on the live check's passive front-door checks? Aggregates only.
- [Penthera 2.1.0](https://github.com/danoszz/penthera/releases/tag/v2.1.0) (Release, 2026-10-29): What changed, in the changelog.
- [E17: the registered run on macOS and Ubuntu made no outside connection and gave the same result, for every command in scope](https://github.com/danoszz/penthera/blob/main/docs/research/prereg/E17-offline-and-determinism.md) (Research, Result, 2026-10-28): URL scans are covered only with --offline: without it, a URL scan downloads the Retire.js database. git runs outside the guard.
- [How we benchmarked Penthera against 25 free tools, losses included](https://penthera.org/blog/sealed-half-benchmark) (Article, 2026-10-28): We tuned on one half of four public datasets and scored once on the other, sealed half. What that method is, what it found, and how far to trust it.
- [We benchmarked 10 free secret scanners on 2,220 labelled secrets. We came 2nd.](https://penthera.org/blog/secret-scanners-creddata) (Article, 2026-10-28): gitleaks, trufflehog, deepsecrets, Nosey Parker and five more on Samsung CredData's sealed half, with intervals, the limits, and where Penthera loses.
- [E14: merging the free tools found more secrets but cost more than 0.02 precision, so --giants keeps Penthera-first](https://github.com/danoszz/penthera/blob/main/docs/research/prereg/E14-giants-combination.md) (Research, Result, 2026-10-25): Does one report from Penthera and the free tools find more than Penthera alone, without losing more than a fixed margin of precision? The answer sets the default for --giants.
- [Leaked secrets](https://penthera.org/benchmarks/secrets) (Benchmark, 2026-10-25): Every tool, every interval, on the sealed half.
- [Malicious npm packages](https://penthera.org/benchmarks/npm-malware) (Benchmark, 2026-10-25): Every tool, every interval, on the sealed half.
- [Malicious PyPI packages](https://penthera.org/benchmarks/pypi-malware) (Benchmark, 2026-10-25): Every tool, every interval, on the sealed half.
- [Injection bugs in JavaScript libraries](https://penthera.org/benchmarks/js-injection) (Benchmark, 2026-10-25): Every tool, every interval, on the sealed half.
- [Penthera 2.0.0](https://github.com/danoszz/penthera/releases/tag/v2.0.0) (Release, 2026-10-25): What changed, in the changelog.
- [E11: Penthera 1.9.0 missed all five registered targets on the three public datasets](https://github.com/danoszz/penthera/blob/main/docs/research/prereg/E11-public-benchmarks.md) (Research, Result, 2026-10-24): One public benchmark per capability, with answer keys we had no hand in: secrets on Samsung CredData, code sinks on SecBench.js, malicious packages on DataDog's dataset.
- [E11b: 2nd on secrets, tied 1st to 3rd on npm and PyPI malware, tied 1st to 4th on JS injection](https://github.com/danoszz/penthera/blob/main/docs/research/prereg/E11b-improvement-study.md) (Research, Result, 2026-10-24): Tune on a dev half, then score a sealed test half once. Only the test half's numbers are published as the after.
- [Penthera 1.9.0](https://github.com/danoszz/penthera/releases/tag/v1.9.0) (Release, 2026-10-23): What changed, in the changelog.
- [eMudhra's broken CA keys show in the public key, and the TLS check now looks for them](https://github.com/danoszz/penthera/blob/main/docs/research/emudhra-weak-keys-2026-10.md) (Incident lesson, 2026-10-22): Each private key was an integer below 256, so a passive handshake is enough to find it. A match is proof, so the finding is confirmed.
- [GitLab's critical patch: a passive scan cannot read a self-managed GitLab's version](https://github.com/danoszz/penthera/blob/main/docs/research/gitlab-version-exposure-2026-10.md) (Incident lesson, 2026-10-22): Asset fingerprinting is ruled out. The note proposes a rule that reads the version from the repository instead; no rule shipped.
- [PHP's 21 Oct flaws: Penthera checks the version a site shows, and says when it cannot tell](https://github.com/danoszz/penthera/blob/main/docs/research/case-study-php-2026-10.md) (Incident lesson, 2026-10-21): Released in 1.8.0. A hidden version is never reported as safe, and the fixed versions match the ones in OpenAI's own checker.
- [E1: Penthera's accuracy on real code](https://github.com/danoszz/penthera/blob/main/docs/research/prereg/E1-real-world-accuracy.md) (Research, Registered, 2026-10-20): How accurate is Penthera's static analysis on real web apps and agent setups its authors did not write? Precision and recall per category, each with a 95% interval.
- [2026 agent incidents: six gaps for Penthera, from verifier integrity to agent sandboxes](https://github.com/danoszz/penthera/blob/main/docs/research/agent-incidents-2026.md) (Incident lesson, 2026-10-20): Agents under evaluation reached real systems they were never meant to touch. Each lesson ends in an action, and the draft study E8 tests one of them.