Skills
Navigate
skills/sandbaseai/sandbase-skills/ai-seo/Gen Agent Trust Hub

ai-seo

Pass

Audited by Gen Agent Trust Hub on Oct 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from the external SandBase API, which could potentially contain malicious instructions or malformed content intended to influence agent behavior.
  • Ingestion points: Responses from sandbase_run and sandbase_run_get as defined in references/sandbase-api-map.md and used in SKILL.md.
  • Boundary markers: No specific delimiters or safety instructions are provided to the agent for isolating the API output from the core system prompt.
  • Capability inventory: The agent can modify the local filesystem (implied SEO artifacts) and continue interacting with the SandBase API (network capability).
  • Sanitization: The skill does not define any logic for filtering, escaping, or validating the structure of external data before interpolation into the agent's context.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 4, 2026, 11:44 AM