skills/megastep/codex-skills/blog-geo/Gen Agent Trust Hub

blog-geo

Pass

Audited by Gen Agent Trust Hub on Feb 26, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface. Ingestion points: Step 1 in SKILL.md extracts content from arbitrary blog posts. Boundary markers: No delimiters or 'ignore embedded instruction' warnings are present to separate user content from agent instructions. Capability inventory: The skill produces text reports and citation capsules; no file-system write access or network operations were identified in the provided files. Sanitization: No input sanitization or validation is performed on the ingested text.
  • [COMMAND_EXECUTION]: The report template in Step 10 of SKILL.md references a command '$blog analyze '. This is documented as a complementary utility from the vendor 'megastep' and is presented as a suggested user action rather than an automated or unauthorized system call.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 26, 2026, 01:29 AM