incident-response-incident-response
Pass
Audited by Gen Agent Trust Hub on Apr 14, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it directly interpolates unvetted external data into instructions for specialized subagents.
- Ingestion points: The
$ARGUMENTSvariable is used across all phases inSKILL.md(e.g., Phase 1, Step 1 and Phase 2, Step 4), which typically contains data from external monitoring systems like PagerDuty or Opsgenie. - Boundary markers: None. The external data is concatenated directly into natural language prompts without using delimiters or instructions to ignore embedded commands.
- Capability inventory: The workflow utilizes a
Tasktool to invoke agents with sensitive capabilities, such asbackend-architect(fix design) anddeployment-engineer(executing production deployments). - Sanitization: None. The skill does not implement any validation, escaping, or filtering for the input data before it is processed by the subagents.
Audit Metadata