AI Stack Alert Critical CVE-2026-119050 For AppSec, AI Eng, CISO, SOC / IR AI-assisted
Agno PandasTools unsafe reflection (CVE-2026-119050) allows code execution via prompt injection
Agno through version 3.1.11 has an unsafe reflection flaw in PandasTools. Model-controlled function names are passed to getattr(), so arbitrary pandas functions can be invoked. Via prompt injection, an attacker can make the agent call pandas.read_pickle on attacker-controlled paths for code execution, or call other readers to read local files. The CVSS base score is 9.2.
Why it matters
Organizations running Agno agents with PandasTools that process untrusted input are exposed. An attacker who can influence the model's input may gain code execution on the host or read local files.
Do this
- Check whether any Agno deployments use PandasTools and are on version 3.1.11 or earlier.
- Upgrade Agno to a fixed release once the maintainers publish one; check the vendor advisory for the fixed version.
- Until patched, remove PandasTools from agents that handle untrusted prompts or external content.
- Restrict agent file system access and run agents in sandboxed, least-privilege environments.
- Review agent logs for unexpected pandas reader calls, especially read_pickle, and for unusual file paths.
Exploit Clock
Exploitation unlikely (for now)
- EPSS 0.43%
Signals
- NoCISA KEV
- 9.2CVSS
- 0.43%EPSS
- NoPublic PoC
- 41/100Priority
- 70%Confidence
Evidence
“Agno through 3.1.11 contains an unsafe reflection vulnerability in PandasTools that allows attackers to invoke arbitrary pandas functions by supplying model-controlled function names to getattr().”
“Attackers can use prompt injection to make the agent call pandas.read_pickle on attacker-controlled paths for code execution, or call other readers to read local files.”
Sources
Affected / related
Agno, PandasTools, pandas, Prompt injection leading to unsafe reflection
Published 2026-10-29 14:17 UTC · updated 2026-11-08 02:32 UTC · AI-assisted summary, checked against quoted sources · How we score