SGLang: Code injection via debug dumper filter in /dumper/configure
Summary
SGLang versions 0.5.10 through 0.5.22 contain a code injection flaw in the debug dumper filter evaluation (_evaluate_filter), which uses eval() with a bypassable sandbox. When the dumper server is enabled via DUMPER_SERVER_PORT, a remote attacker can send a crafted filter expression to the /dumper/configure endpoint and execute Python code in the scheduler process on the next forward pass. The CVE describes the endpoint as unauthenticated, but the research indicates that a valid API key may be sufficient, so this point is disputed.
Affected
- SGLang 0.5.10 through 0.5.22
- Only deployments with the debug dumper server enabled (DUMPER_SERVER_PORT set)
- Versions before 0.5.10: Unknown.
Fixed in
Unknown. No vendor-confirmed fixed release found. Fix work is in open pull requests (#45998, #44782); their merge status is unclear.
Workaround
- Ensure DUMPER_SERVER_PORT is unset in every deployment
- Restrict network access to the SGLang API to trusted hosts; do not expose it to the internet
- Block the /dumper/ routes at a reverse proxy or firewall if the dumper must remain enabled
- No vendor mitigation published.
Exploitation
- Public proof of concept: yes, documented in GitHub issue #44188
- In the wild: No evidence found
- CISA SSVC: exploitation PoC, automatable: no, technical impact: total
- EPSS is low (0.00452)
- Not listed in CISA KEV (per research)
Our recommendation
- Inventory all SGLang installations, including PyPI sglang packages
- Check whether DUMPER_SERVER_PORT is set and disable the dumper if not needed
- Restrict access to SGLang endpoints to trusted hosts immediately
- Monitor the GitHub issue and pull requests and upgrade once a fixed release is published
- Review the scheduler host for unexpected files or processes if exposure existed
Public exploits
PoC-in-GitHub, Metasploit and Exploit-DB list nothing for this CVE.
Research sources
- CVE record: VulnCheck advisory www.vulncheck.com
- GitHub issue #44188 github.com
- dumper.py at v0.5.22 (vulnerable code) github.com
- SGLang repository github.com
- Pull request #45998 (eval replacement) github.com
- Pull request #44782 (hardening attempt) github.com
- VulDB entry 421267 vuldb.com
- SUSE CVE record via Vulners vulners.com
Researched and written up in English and German by an AI (Claude by Anthropic) based on NVD, CISA, EPSS and the sources listed. The vendor information is authoritative. Researched on 31 Oct 2026, 04:22.
Original description (NVD)
SGLang 0.5.10 through 0.5.22 contains a code injection vulnerability in _evaluate_filter() that allows remote attackers to execute Python code by escaping the eval() sandbox through re.search __globals__. Attackers can send a crafted filter expression to the unauthenticated /dumper/configure endpoint when DUMPER_SERVER_PORT is set, running code in the scheduler process on the next forward pass.
NVD references (5)
- https://github.com/sgl-project/sglang
- https://github.com/sgl-project/sglang/blob/v0.5.22/python/sglang/srt/debug_utils/dumper.py#L1309-L1320
- https://github.com/sgl-project/sglang/blob/v0.5.22/python/sglang/srt/debug_utils/dumper.py#L1380-L1401
- https://github.com/sgl-project/sglang/issues/44188
- https://www.vulncheck.com/advisories/sglang-0.5.10-through-0.5.22-code-injection-via-debug-dumper-filter
Sources
The tracker combines official, freely usable data sources – updated continuously:
- NVD (NIST) – CVEs, rating (CVSS), affected products
- CISA KEV – vulnerabilities confirmed by CISA as actively exploited
- CISA Vulnrichment – CISA assessment (SSVC) in the official CVE record
- FIRST EPSS – probability of exploitation
- Microsoft MSRC – Patch Tuesday: vulnerabilities, severity, updates
- PoC-in-GitHub – public proof-of-concept repositories
- Metasploit (Rapid7) – modules of the Metasploit Framework
- Exploit-DB (OffSec) – archive of public exploits (links only)
- Patch days: official vendor pages (linked per vendor) (Vendor patch days)
This product uses data from the NVD API but is not endorsed or certified by the NVD. EPSS data © FIRST.org.
Affected? We can help.
We check whether your systems are affected, install updates and close the gap – continuously with the System Health Checker if you wish.
