xmpp-enumeration

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs the agent to embed plaintext passwords/credentials verbatim into commands and XML (e.g., SASL PLAIN base64, in-band registration XML), includes examples of hardcoded passwords and Write("/tmp/...","") usage, and requires reporting "new credentials or tokens found" — all of which force the LLM to handle and output secret values directly.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.85). The skill explicitly documents deliberate user-enumeration and credential-harvesting techniques (in-band registration conflict enumeration, automated account creation, SASL auth probing, and MUC history retrieval) that enable credential theft and data extraction and can be readily abused outside an authorized testing context.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow performs XMPP enumeration by connecting to the target server and then ingesting server responses (XML stream features, IQ results, roster/disco/search results, and MUC history) into the agent’s LLM context; these responses are outsider-authored content from the remote XMPP service (not authored by the operating user).