alert-triage
Pass
Audited by Gen Agent Trust Hub on Jul 20, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data such as alert payloads and log entries. This creates a surface for indirect prompt injection, which the skill proactively addresses through the 'Prompt Injection Safety Notice' in Section 8, instructing the agent to ignore any embedded directives.
- [PROMPT_INJECTION]: The use of '$ARGUMENTS' allows external input to define the investigation scope. The skill includes hardening instructions to ensure these inputs are treated as data rather than instructions, maintaining the agent's behavioral integrity.
- [SAFE]: All external references point to official and trusted security framework repositories (NIST and MITRE). No unauthorized downloads or remote script executions are performed.
- [SAFE]: The agent is limited to a non-intrusive toolset (Read, Grep, Glob), ensuring it cannot perform unauthorized network communication or system modifications.
Audit Metadata