Slack General Query

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were detected in the skill's instructions or metadata.
  • [NO_CODE]: The skill consists exclusively of configuration and conversational guidelines. It does not include any scripts, executable commands, or external package dependencies.
  • [PROMPT_INJECTION]: Analysis of the indirect prompt injection surface identifies the following:
  • Ingestion points: The skill receives untrusted user input via the raw_query parameter.
  • Boundary markers: No specific boundary markers are utilized in the prompt template.
  • Capability inventory: The skill requests no capabilities (empty requires_capabilities), meaning it cannot perform file system, network, or command-line operations.
  • Sanitization: No sanitization is performed on the input string, though the lack of capabilities renders this surface harmless.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 11:37 AM