google/skills application-integration-troubleshooting
Checked the catalogued skill at application-integration-troubleshooting.
D; retained sample includes a high finding · SD-007 · what this rule accepts as a false positive →
F; retained sample includes a high finding · SD-003 · what this rule accepts as a false positive →
A set by no findings on this axis
8 suppressed by AI triage
The file is an agent skill manifest containing a declarative warning, not executable malicious code.
The snippet is a documentation example of a curl command, not an active malicious payload.
The content does not contain an actual hardcoded API key; only a generic mention of the sk- prefix.
No path traversal pattern (e.g., ../ or absolute system paths) appears in the snippet.
Placeholder config file with no actual outbound network call.
The matched text is a documentation note about detecting DNS exfiltration, not actual malicious activity or code.
Same reasoning: documentation example, not a threat.
The matched text is a placeholder indicating redaction, not an actual hardcoded secret.