content-ideas
Pass
Audited by Gen Agent Trust Hub on Jul 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the
kai-reporttool to retrieve keyword data in JSON format for analysis and useskai-configfor configuration management. - [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection because it ingests and processes search queries (keywords) from Google Search Console which are controlled by external searchers.
- Ingestion points: External keyword data retrieved via the
kai-reporttool inSKILL.md(Step 1). - Boundary markers: Absent; the skill does not utilize delimiters or specific instructions to the agent to ignore instructions potentially embedded within the keyword strings.
- Capability inventory: The skill utilizes shell command execution via
kai-reportandkai-config(SKILL.md Step 1 and Error Handling sections). - Sanitization: Absent; there is no evidence of filtering, escaping, or validation of the search query strings before they are processed by the agent for scoring and ranking.
Audit Metadata