breach-patterns

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a documentation-only resource that provides educational patterns for security audits based on public breach disclosures. It does not contain any executable scripts, obfuscated code, or hidden malicious instructions.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines an attack surface for indirect prompt injection through the use of web-retrieval tools.
  • Ingestion points: The skill encourages the use of WebSearch and WebFetch to ingest content from external security blogs, post-mortems, and news sites.
  • Boundary markers: There are no explicit instructions for the agent to use delimiters or ignore embedded instructions when processing data fetched from external URLs.
  • Capability inventory: The agent has access to Bash, Read, Grep, Glob, WebSearch, and WebFetch tools.
  • Sanitization: The instructions do not specify any validation or sanitization steps for content retrieved from external sources before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 02:07 AM
Security Audit — agent-trust-hub — breach-patterns