suparank/optimize
Pass
Audited by Gen Agent Trust Hub on Feb 23, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill ingests untrusted article content and has the capability to modify the filesystem, creating a surface for indirect instructions to influence agent behavior.
- Ingestion points: Article content is read from
.claude/suparank-content/[folder]/article.mdand templates from~/.claude/skills/suparank/templates/. - Boundary markers: The skill instructions do not define delimiters or specific 'ignore' instructions when processing external article content.
- Capability inventory: The agent can modify and save files (
edit the article file directly and re-save). - Sanitization: No sanitization or validation of the ingested article content is performed before processing or editing.
Audit Metadata