social-performance-review-codex

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external, potentially untrusted data to generate reports and update project state.
  • Ingestion points: The skill accepts CSV exports, platform summaries, and reporting spreadsheets from various third-party services like Instagram, LinkedIn, and scheduling tools (SKILL.md).
  • Boundary markers: The instructions do not specify any delimiters or safety prompts to prevent the agent from following instructions embedded within the social media data.
  • Capability inventory: The skill has the authority to write to outputs/reviews/ and update persistent context files such as context/best-performers.md and context/review-history.md (SKILL.md).
  • Sanitization: There is no mention of data sanitization, schema validation, or filtering of the input data before it is processed and stored.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 10:59 AM
Security Audit — agent-trust-hub — social-performance-review-codex