identify-kill-list

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructions define a workflow for analyzing data from sources like audience notes and portfolio assessments, which presents an injection surface.
  • Ingestion points: Data is pulled from section-portfolio-assessment and audience-notes contexts as specified in the workflow.
  • Boundary markers: The skill lacks specific delimiters or guardrails to treat external data as untrusted content.
  • Capability inventory: There are no shell commands, file-write operations, or network calls identified within the skill body or references.
  • Sanitization: No input validation or content filtering is prescribed for the analyzed data.
  • [NO_CODE]: The skill consists entirely of markdown text and natural language instructions with no accompanying scripts, binaries, or tool-invocation logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 02:37 PM
Security Audit — agent-trust-hub — identify-kill-list