oa-system-attack

SUSPICIOUS/HIGH-RISK offensive skill. The content is internally consistent with its stated purpose, but that purpose is to equip an AI agent with intrusion and exploitation capabilities against OA/internal systems, including credential harvesting after compromise. No hidden installer or covert third-party credential relay is evident, so this is not confirmed malware, but it is a dangerous exploit-oriented skill with substantial real-world abuse potential.

This fragment is an explicit offensive exploitation and automation playbook targeting real enterprise OA products and related internal services. It contains multiple weaponized RCE paths (webshell/eval, SSRF→JNDI→RCE, deserialization/XMLDecoder→Runtime.exec), credential/secret theft mechanisms (config/properties retrieval, SQLi extraction, default credential usage), and destructive capability (arbitrary file delete). It is not legitimate dependency/library code and should be treated as highly malicious content with extreme security risk if present in a software supply-chain artifact.

This fragment is unequivocally offensive exploit/playbook content, detailing end-to-end attack chains to compromise enterprise OA systems via session hijacking, RCE (including webshell upload and interpreter/deserialization execution), arbitrary file read/write via traversal, SQL injection credential/config extraction, and SSRF. It is not legitimate dependency/library code. If included in any distributed package, it represents a severe supply-chain security risk due to its direct enablement of system compromise.

No executable malware logic is present in the provided fragment; however, it is highly likely to be malicious content (a weaponized prompt/expected-output specification) that explicitly guides unauthorized compromise and RCE/webshell/shell acquisition for real enterprise OA products. Any system that consumes this artifact to drive automated behavior should treat it as high risk despite the lack of direct code execution.