Skip to content
KitploitKITPLOIT
ToolsExploitsBlog
Log in
Submit
ToolsExploitsBlog
Submit

Hacking, PenTest, and Cybersecurity Tools for Your Security Arsenal!

Kitploit is a directory of hacking, cybersecurity, and pentesting tools. Discover the latest project updates to find vulnerabilities, analyze systems, automate testing, and strengthen your security.

FeedsContactPrivacy© 2026 Kitploit

Tool Directory

Categories

View all categories
Loading categories
shiro-check — 扫出你实际装的 Apache Shiro 模块与版本,逐条判定官方 26 条 CVE 里哪些真的落在你身上。按「CVE × 模块」判定,零依赖单 jar。 CVE-2026-49268 | Kitploit
Tools/GitHubGitHub/xiaoqimikko/shiro-check
Static AnalysisVulnerability ScannersVulnerability AnalysisDevSecOpsSupply Chain Security
GitHubxiaoqimikko/shiro-check

shiro-check

扫出你实际装的 Apache Shiro 模块与版本,逐条判定官方 26 条 CVE 里哪些真的落在你身上。按「CVE × 模块」判定,零依赖单 jar。 CVE-2026-49268

View Repository
152 months agoNot yet reviewed
Share

shiro-check

Scan the Apache Shiro modules and versions you actually have installed, and determine one by one which of the 26 official CVEs truly affect you.

Zero-dependency single jar, no network access, does not read your pom — it directly scans build artifacts (jar / war / fat jar / directory).

java -jar shiro-check.jar --utf8 your-app.jar

Why "just glance at the version number and call it done" doesn't work

Shiro's CVEs are not attached to "Shiro" as a whole — they're attached to specific modules. The same CVE ID means different rules on different modules:

CVEAffected modulesPeople using only shiro-core
CVE-2020-17510shiro-springNot affected
CVE-2020-17523shiro-web / shiro-spring / shiro-spring-boot-starterNot affected
CVE-2023-34478shiro-webNot affected
CVE-2026-56091shiro-guiceNot affected

Of the official 26 entries, 13 don't touch shiro-core at all. Compressing them into a single line like "Shiro < 1.7.1 has vulnerabilities" means making the judgment the user should make for themselves — and making it wrong.

This tool's granularity is CVE × module: the 26 CVEs expand into 37 rules covering 7 modules.

Three things coordinate-based matching can't see

1. Five entries never make it into Dependabot alerts

Dependabot 按依赖坐标匹配 GitHub advisory。26 条里有 5 条对不上:

条目为什么对不上
CVE-2026-56091(shiro-guice 认证绕过,high)advisory 仍是 unreviewed,受影响包为空
CVE-2026-56130(RememberMe cookie 永不过期)同上
CVE-2014-0074(LDAP 空口令绕过)同上
CVE-2023-22602(Spring Boot 2.6+ 认证绕过,high)advisory 只挂在 org.apache.shiro:shiro-root 上
CVE-2010-3863同上

And shiro-root is a packaging=pom parent POM — on Maven Central there is no corresponding jar (HTTP 404), no project would ever depend on it, so coordinate-based matching can never line up.

ℹ️ unreviewed is a normal workflow state on GitHub (NVD auto-import, affected packages not yet manually annotated), and coordinate-based matching is a reasonable design too. This only states the fact that "they never make it into alerts" — it's not saying anyone reported it wrong.

2. The uber jar hides other modules inside

shiro-all is an uber jar that bundles multiple modules together. Testing shows that inside shiro-all-1.3.2.jar there are 6 META-INF/maven/org.apache.shiro/<module>/pom.properties files:

shiro-all  shiro-core  shiro-web  shiro-spring  shiro-ehcache  shiro-quartz

That is, a project that only declares shiro-all in its pom has only one name at the coordinate level (among the 26 entries, only CVE-2016-6802 is tied to this coordinate), while the jar actually contains the code of three modules: core / web / spring.

Scanning the actual artifact can see through this layer. In a real test with a shiro-all-1.3.2.jar, this tool reported 20 entries across 3 modules.

3. Three entries have official fixed versions that are unavailable on Maven Central

The first_patched_version of 3 advisories is 3.0.0-alpha-2, and this version was never published to Maven Central (the official release went straight to 3.0.0 final). Upgrading according to it is impossible — this tool will flag this and replace the upgrade recommendation with a version that is actually obtainable.

A finding does not equal "you are compromised"

"Hit" = this module is present and the version falls within the official affected range. It does not equal "already exploited" or "necessarily exploitable".

The vast majority of the 26 entries require specific configurations to hold, so the tool lists them separately:

  • 🔴 Affected under default configuration — handle first (e.g., CVE-2026-43827 session fixation)
  • ⚠️ Requires specific conditions — conditions are given for each entry; if not met, you are not affected

Examples of conditions (compared verbatim against the original official descriptions, with no extrapolation):

CVE-2026-49268  仅当使用 DefaultLdapRealm(用户名未转义即拼进 LDAP DN)
CVE-2023-22602  仅当与 Spring Boot 2.6+ 一起使用,且未把
                spring.mvc.pathmatch.matching-strategy 设回 ant_path_matcher
CVE-2026-23903  仅当静态文件放在大小写不敏感的文件系统上(如 macOS 默认设置),
                且 Shiro 里只配了小写的 filter 路径;只影响静态文件

🔴 The tool does not parse your configuration. Shiro can be configured in shiro.ini / application.yml / Java code, and a parsed conclusion is more dangerous than no parsing at all. The conditions are listed in plain text for you to judge yourself — this is intentional design, not laziness.

Usage

# 扫一个 jar / war
java -jar shiro-check.jar your-app.jar

# 扫整个目录(递归)
java -jar shiro-check.jar /path/to/libs

# 连「不适用」的条目也列出来
java -jar shiro-check.jar --all your-app.jar

# Windows 控制台中文乱码时
java -jar shiro-check.jar --utf8 your-app.jar

Detection capability: plain jar · Spring Boot fat jar (BOOT-INF/lib/) · traditional WAR (WEB-INF/lib/) · uber jar (shiro-all, with contained modules expanded) · renamed jars (determined by pom.properties) · malformed archives with backslash paths.

Requires JDK 17+. Zero runtime dependencies.

How the decision table was produced

Not a single line is copied by hand. tools/gen_rules.py generates it from two primary sources:

源提供什么
shiro.apache.org/security-reports.html条目全集、描述原文、触发条件原文
GitHub Advisory API受影响模块坐标、结构化版本区间、评级

The generation process carries 12 assertions; if any one fails, it aborts without writing the file (to prevent "parsing fails, an empty shell table is generated, and the tests still all pass"):

  • Completeness: the h3 headings in the body and the directory anchors must count out the same set of CVEs
  • Provenance: for manually filled version ranges, the version strings must appear verbatim in the official source text
  • Empirical verification: module attribution must be confirmed by class locations in the real jar ("I remember this one is web" is not accepted)
  • Executability: every fixed version is probed once on Maven Central; those that cannot be obtained are flagged
  • Claims: the number of Dependabot blind spots, cross-module differences, and modules contained in the uber jar — all must still hold

Before publishing, there is also tools/recheck_before_publish.py, which re-verifies the load-bearing evidence using a second, independent implementation (it does not import the generation script, nor read its output — reusing the same parsing logic would reproduce the bugs along with it).

Limitations

  • Covers only the 26 entries on the official security-reports page; does not include undisclosed or third-party integration issues
  • Version ranges are taken from the official source and GitHub; when the two occasionally differ, the structured range on the module coordinates prevails
  • Configuration is not parsed, so conditional entries require you to judge for yourself whether they apply
  • The decision table is a snapshot at generation time; GitHub may at any time turn unreviewed into reviewed

English

shiro-check scans your build artifacts (jar / war / fat jar / directory) to find which Apache Shiro modules you actually ship, then evaluates all 26 official CVEs against your exact module + version combination.

Why it exists: Shiro CVEs attach to modules, not to "Shiro". 13 of the 26 never touch shiro-core at all. The tool works at CVE × module granularity (37 rules, 7 modules).

Three things coordinate-based matching cannot see:

Download Tool

Most Popular

View all →

Discover the most used tools by our community.

  • 1

    OsintHunter

    letox4/osinthunter

    🎯 The ultimate OSINT toolkit: investigation graph, confidence scoring, alias families, map, timeline and reports. 49 modules, 400+ sites, no API keys required.

    795
  • 2

    ghostlock-app

    yukonga/ghostlock-app

    GhostLock One-Tap Execution App (CVE-2026-43499)

    520
  • 3

    ARTEX

    baarai/artex

    LLM multi-agent autonomous penetration testing platform with planner/worker agents, asset and exploration graphs, MITM traffic recording, and vulnerability retesting.

    439
  • 4

    redforge

    cybersecurity91/redforge

    Browser-side red team toolkit with 102 offline tools for recon, payload generation, credential attacks, crypto, encoding and engagement reporting.

    346
  • 5

    looking-glass

    anonghosty/looking-glass

    CSIRT platform that ingests Shadowserver reports into MongoDB and replays them as interactive maps, analysis, findings, advisories and executive dashboards.

    267
  • 6

    CyberStrikeAI

    ed1s0nz/cyberstrikeai

    The system of action for AI-native cybersecurity—where intent becomes governed execution, evidence becomes operational memory, and every operation improves the next.

    264

Explore all tools

Browse our collection of tools

View all tools →