page-strategy
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill acts as a normalization layer that ingests external data to produce artifacts consumed by downstream processes.\n
- Ingestion points: The skill reads external data from
workspace/brand/extract.mdand user-provided briefs as specified in SKILL.md.\n - Boundary markers: There are no specific delimiters or instructions to treat input content as untrusted data or to ignore embedded instructions within the source files.\n
- Capability inventory: The skill writes to
workspace/pages/[page-name]/strategy.mdandstrategy.json. These artifacts are explicitly intended to be read by subsequent skills in the pipeline (e.g.,/page-copy), potentially facilitating multi-step injection chains.\n - Sanitization: The skill lacks mechanisms to sanitize or escape data before interpolation into the strategy artifacts, relying only on content-specific filtering for compliance niches.
Audit Metadata