SWE-agent

!warning Most of our current development effort is on mini-swe-agent, which has superseded SWE-agent. It matches the performance performance of SWE-agent, while being much simpler. See the FAQ for more details about the differences. Our general recommendation is to use mini-SWE-agent instead of...

CVE-2026-108592

mini-swe-agent 1.10.0 through 2.4.6 contains an information exposure vulnerability in BubblewrapEnvironment because bwrap omits --clearenv, so sandboxed commands inherit the host environment. Attackers using prompt injection in processed task content can make the agent read API keys from the...

CVE-2026-108592

The mini-swe-agent software (versions 1.10.0 through 2.4.6 ) is vulnerable to an information exposure flaw within the BubblewrapEnvironment component. The root cause is that the underlying bwrap tool omits the --clearenv flag, causing sandboxed commands to inherit the host's environment variables...

CVE-2026-108592 mini-swe-agent 1.10.0 through 2.4.6 Environment Exposure via BubblewrapEnvironment

mini-swe-agent 1.10.0 through 2.4.6 contains an information exposure vulnerability in BubblewrapEnvironment because bwrap omits --clearenv, so sandboxed commands inherit the host environment. Attackers using prompt injection in processed task content can make the agent read API keys from the...

EUVD-2026-96402

mini-swe-agent 1.10.0 through 2.4.6 contains an information exposure vulnerability in BubblewrapEnvironment because bwrap omits --clearenv, so sandboxed commands inherit the host environment. Attackers using prompt injection in processed task content can make the agent read API keys from the...

CVE-2026-108592 mini-swe-agent 1.10.0 through 2.4.6 Environment Exposure via BubblewrapEnvironment

mini-swe-agent 1.10.0 through 2.4.6 contains an information exposure vulnerability in BubblewrapEnvironment because bwrap omits --clearenv, so sandboxed commands inherit the host environment. Attackers using prompt injection in processed task content can make the agent read API keys from the...

PT-2026-109732

Name of the Vulnerable Software and Affected Versions mini-swe-agent versions 1.10.0 through 2.4.6 Description An information exposure issue exists in the BubblewrapEnvironment where the bwrap tool is executed without the --clearenv flag. This causes sandboxed commands to inherit the host...

CVE-2026-108592: Cleartext Storage of Sensitive Information in an Environment Variable

mini-swe-agent 1.10.0 through 2.4.6 contains an information exposure vulnerability in BubblewrapEnvironment because bwrap omits --clearenv, so sandboxed commands inherit the host environment. Attackers using prompt injection in processed task content can make the agent read API keys from the...

CVE-2026-75482

SWE-agent's trajectory inspector sweagent inspector, confirmed in v1.1.0, is an HTTP server that joins request paths to the trajectory directory in its /trajectory/ handler without rejecting parent-directory '..' references, bypassing the built-in path sanitization. The server binds all interface...

EUVD-2026-60394

SWE-agent's trajectory inspector sweagent inspector, confirmed in v1.1.0, is an HTTP server that joins request paths to the trajectory directory in its /trajectory/ handler without rejecting parent-directory '..' references, bypassing the built-in path sanitization. The server binds all interface...

CVE-2026-75482 SWE-agent Trajectory Inspector Path Traversal File Disclosure

SWE-agent's trajectory inspector sweagent inspector, confirmed in v1.1.0, is an HTTP server that joins request paths to the trajectory directory in its /trajectory/ handler without rejecting parent-directory '..' references, bypassing the built-in path sanitization. The server binds all interface...

CVE-2026-75482 SWE-agent Trajectory Inspector Path Traversal File Disclosure

SWE-agent's trajectory inspector sweagent inspector, confirmed in v1.1.0, is an HTTP server that joins request paths to the trajectory directory in its /trajectory/ handler without rejecting parent-directory '..' references, bypassing the built-in path sanitization. The server binds all interface...

CVE-2026-75482 SWE-agent Trajectory Inspector Path Traversal File Disclosure

SWE-agent's trajectory inspector sweagent inspector, confirmed in v1.1.0, is an HTTP server that joins request paths to the trajectory directory in its /trajectory/ handler without rejecting parent-directory '..' references, bypassing the built-in path sanitization. The server binds all interface...

CVE-2026-75482

SWE-agent's trajectory inspector (v1.1.0) exposes a path-traversal vulnerability in its /trajectory/ handler, allowing unauthenticated clients to read arbitrary files shaped as trajectory JSON, potentially leaking repository contents, command output, and API keys. The server binds to 0.0.0.0 with...

Sifting the Noise: A Comparative Study of LLM Agents in Vulnerability False Positive Filtering

Static Application Security Testing SAST tools are essential for identifying software vulnerabilities, but they often produce a high volume of false positives FPs, imposing a substantial manual triage burden on developers. Recent advances in Large Language Model LLM agents offer a promising...

LLM Agents for Automated Web Vulnerability Reproduction: Are We There Yet?

Large language model LLM agents have demonstrated remarkable capabilities in software engineering and cybersecurity tasks, including code generation, vulnerability discovery, and automated testing. One critical but underexplored application is automated web vulnerability reproduction, which...