exploit-development

Warn

Audited by Socket on Sep 19, 2026

7 alerts found:

Securityx4Anomalyx3
SecurityMEDIUM
SKILL.md

SUSPICIOUS. The skill is a coherent exploit-development/offensive security guide, but that purpose itself gives the agent high-risk offensive capabilities: exploit creation, shellcode generation, remote delivery, defense evasion, and even kernel exploit examples. Install provenance is mostly legitimate and not the main issue; the core risk is enabling an AI agent to perform exploitation workflows with real-world impact.

Confidence: 97%Severity: 93%
SecurityMEDIUM
test-cases.md

The fragment is offensive security and exploit-development documentation, not direct malware code. It contains no visible autonomous exfiltration, persistence, credential theft, or package-install execution. Its content is highly actionable and includes remote exploitation and shell/root-shell objectives, so use is appropriate only in authorized laboratory or assessment environments. The principal concern is misuse rather than supply-chain malware in this fragment.

Confidence: 96%Severity: 72%
AnomalyLOW
guides/shellcode-writing-encoding-guide.md

The fragment is exploit-development and shellcode training material. It contains explicit techniques for generating, encoding, and executing /bin/sh payloads, so it presents a meaningful misuse and execution risk if incorporated into tooling or run with untrusted input. It does not itself demonstrate package malware, data theft, or a supply-chain backdoor; the observed behavior is documented dual-use functionality.

Confidence: 98%Severity: 62%
AnomalyLOW
guides/pwntools-exploit-development-guide.md

The fragment contains high-impact exploit-development examples capable of hijacking a vulnerable process, spawning shells, exposing a bind shell, and reading a flag file. Those capabilities create a substantial misuse risk, but the code shown does not exhibit covert malicious package behavior or supply-chain malware indicators. Assessment is limited to the visible fragment and its apparent educational/CTF context.

Confidence: 97%Severity: 68%
SecurityMEDIUM
payloads.md

This fragment is an offensive security and exploit-development tutorial section. It contains code and commands capable of spawning shells, corrupting memory, executing shellcode, and modifying privileged files when deliberately compiled or run against suitable targets. It does not itself demonstrate automatic malware behavior, data exfiltration, persistence, or a package supply-chain backdoor. Use should be restricted to authorized labs and vulnerable test systems.

Confidence: 98%Severity: 70%
SecurityMEDIUM
guides/buffer-overflow-to-rop-chain-guide.md

This fragment is an explicit stack-buffer-overflow/ROP exploit that leaks libc memory and spawns an interactive shell through system('/bin/sh'). It should only be used in authorized CTF, lab, or security-testing environments. It is not obfuscated and shows no covert exfiltration or persistence, but its direct shell-spawning behavior creates a high security risk if applied to an unintended target.

Confidence: 99%Severity: 95%
AnomalyLOW
guides/heap-exploitation-guide.md

The fragment is exploit-development and heap-security educational material. It presents techniques that could be used to exploit vulnerable native programs and includes an exercise targeting shell execution, but it does not itself contain executable malware, data theft, persistence, or unauthorized network activity. The primary concern is dual-use offensive guidance rather than supply-chain malware.

Confidence: 98%Severity: 55%
Audit Metadata
Analyzed At
Sep 19, 2026, 04:40 PM
Package URL
pkg:socket/skills-sh/brucesongs%2Fkali-claw%2Fexploit-development%2F@4bdc67e6b78d5e6691b9567c6bb0523610da4c7ac154acb4c662ad75faf88106