huaweicloud/huaweicloud-skills huawei-cloud-skill-audit
Checked the catalogued skill at huawei-cloud-skill-audit.
F; retained sample includes a critical finding · SD-002 · what this rule accepts as a false positive →
F; retained sample includes a critical finding · SD-004 · what this rule accepts as a false positive →
A set by no findings on this axis
18 suppressed by AI triage
The content is a configuration rule for detecting outbound curl calls, not an actual outbound network call.
The content is a configuration rule for detecting curl basic auth usage, not an actual outbound network call.
The URL is a configured LLM endpoint used by the script, not a covert exfiltration target.
The script imports urllib to make a legitimate outbound call to an LLM API, not a malicious payload fetch.
The default base URL points to a known LLM service used by the application.
The code explicitly performs a normal API request to the LLM endpoint, a standard operation.
The urllib call is part of the normal LLM request flow and not an unauthorized exfiltration.
Line 144 is a comment describing expected behavior, not executable persistence code.
Line 146 is a comment referencing a path; it does not execute or write to the file system.
The flagged string does not contain any credential paths; it appears to be a generic comment.
The file is a documentation snippet and contains no executable code that performs path traversal
No credential access indicators; generic redacted string matched.
The file is documentation and contains no executable path traversal code.
The file contains no actual path traversal patterns, just empty lines; likely a false positive.
No credential path references are present; the match is likely a false positive.
Documentation example, no real path traversal risk.
The file is documentation and contains no actual path traversal strings.
The text appears to be a documentation placeholder rather than an actual outbound network call or configuration.