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
CVE-2026-9147-uproot-rce — uproot <= 5.7.4 code injection via unsafe Python source generation from ROOT TStreamerInfo metadata. | Kitploit
Tools/GitHubGitHub/saiteja-erukude/cve-2026-9147-uproot-rce
Payload GenerationVulnerability AnalysisExploitation
GitHubsaiteja-erukude/cve-2026-9147-uproot-rce

CVE-2026-9147-uproot-rce

uproot <= 5.7.4 code injection via unsafe Python source generation from ROOT TStreamerInfo metadata.

View Repository
Share
361 month agoNot yet reviewed

CVE-2026-9147: uproot Code Injection via TStreamerInfo Metadata

Severity: High, CVSS 4.0 8.5, CVSS 3.1 7.8 (assigned by VulnCheck, the CNA)

Vector (v4.0): CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N

Vector (v3.1): CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

Affected: uproot <= 5.7.4

Fixed in: 5.7.5 / commit c045c28

CWE: CWE-94 (Improper Control of Generation of Code, Code Injection); secondary CWE-95 (Eval Injection)

Discovered and reported by: Sai Teja Erukude

Published: July 18, 2026


Summary

uproot 5.7.4 and prior dynamically generate Python class source from ROOT TStreamerInfo metadata and compile it at runtime.

Some file-controlled streamer metadata fields are interpolated into generated Python source without safe quoting. An attacker who can provide a crafted ROOT file can inject Python expression-breaking content into streamer metadata, such as a crafted element name.

When an affected application opens the crafted ROOT file and uproot generates and invokes the corresponding reader method, the injected Python expression can execute in the context of the hosting process.

Impact

An attacker may execute arbitrary Python code in applications that open or process attacker-controlled ROOT files with affected uproot code paths.

Depending on the application, this can allow:

  • Reading or modifying files accessible to the process.
  • Accessing environment variables and application secrets.
  • Calling internal services reachable from the host.
  • Disrupting data-processing jobs or application behavior.

Technical Detail

The issue is in uproot's generated-source path for ROOT streamer metadata.

The PoC creates a ROOT file containing crafted TStreamerInfo metadata for an ExploitTarget class. The crafted streamer element name breaks out of the generated expression and injects a benign pathlib call.

The reported behavior is:

  1. uproot opens the crafted ROOT file.
  2. uproot reads file-backed TStreamerInfo metadata.
  3. uproot generates Python class source for ExploitTarget.
  4. The generated class source contains the injected expression.
  5. Invoking the generated read_members method executes the injected expression.

Proof of Concept

root_file_builder.py creates malicious_tstreamerinfo.root with crafted streamer metadata.

run_poc.py opens the file with uproot, forces generation of the file-backed ExploitTarget class, invokes the generated read_members method, and verifies marker-file evidence.

Run in a local test environment only:

python -m venv .venv
.venv\Scripts\activate
python -m pip install -r requirements.txt
python -B run_poc.py

Expected evidence on vulnerable versions:

uproot_version: 5.7.4
file_backed_element_name_contains_payload: True
class_code_contains_injected_pathlib_call: True
expected_exception: True
exception_type: DeserializationError
marker_exists: True
marker_contents:
executed through uproot generated read_members
success: True

The payload is intentionally harmless. It only writes this local marker string:

executed through uproot generated read_members

It does not spawn a shell, connect to a network service, read secrets, delete data, or modify files outside the PoC directory.

Remediation

Upgrade to uproot 5.7.5 or later.

If immediate upgrade is not possible:

  • Do not process ROOT files from untrusted users.
  • Isolate ROOT-file processing jobs with minimal privileges.
  • Validate or reject unexpected TStreamerInfo metadata before processing.
  • Avoid workflows that automatically open uploaded or externally supplied ROOT files.

Disclosure Timeline

DateEvent
May 20, 2026Vulnerability submitted to VulnCheck
May 20, 2026VulnCheck initiated outreach; CVE-2026-9147 provisionally allocated
July 5, 2026uproot 5.7.5 released with patch
July 18, 2026CVE-2026-9147 published

Credit

Discovered and reported by Sai Teja Erukude.

Disclosure was coordinated through VulnCheck.

References

  • CVE Record: https://www.cve.org/CVERecord?id=CVE-2026-9147
  • NVD: https://nvd.nist.gov/vuln/detail/CVE-2026-9147
  • VulnCheck advisory: https://www.vulncheck.com/advisories/uproot-and-before-code-injection-via-tstreamerinfo-metadata
  • GitHub advisory: https://github.com/scikit-hep/uproot5/security/advisories/GHSA-6946-mq52-g438
  • Patch commit: https://github.com/scikit-hep/uproot5/commit/c045c2824295d907d2e705f31110c742928e50e7
  • uproot project: https://github.com/scikit-hep/uproot5
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.

    786
  • 2

    ghostlock-app

    yukonga/ghostlock-app

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

    506
  • 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.

    429
  • 4

    redforge

    cybersecurity91/redforge

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

    345
  • 5

    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.

    262
  • 6

    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.

    256

Explore all tools

Browse our collection of tools

View all tools →