escalation-pack
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to summarize and process untrusted external data such as customer logs and ticket descriptions.
- Ingestion points: Identified in
SKILL.md(Workflow Step 3) andreferences/escalation-pack-checklist.md(Evidence and Required Fields sections). - Boundary markers: There are no instructions to use delimiters or ignore instructions embedded within the processed data.
- Capability inventory: The skill does not define any custom tools, scripts, or network capabilities, significantly reducing the impact of potential injection attacks.
- Sanitization: The instructions include proactive requirements for redacting passwords, tokens, and personally identifiable information (PII).
- [NO_CODE]: The skill consists exclusively of markdown instructions and YAML configuration files, containing no executable scripts or external dependencies.
Audit Metadata