content-audit

Pass

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted user-provided content (drafts and brand guidelines) without implementing explicit boundary markers or sanitization. An attacker could embed instructions within a draft to manipulate the audit scoring or influence the agent's behavior during the audit process.
  • Ingestion points: User-provided draft text and brand reference files (brand/VOICE.md, brand/VISUAL.md) mentioned in SKILL.md.
  • Boundary markers: Absent; the skill does not specify delimiters or system instructions to ignore embedded commands within the analyzed content.
  • Capability inventory: The skill's primary capabilities are text analysis, scoring, and providing feedback. It does not perform file writes, network operations, or subprocess executions.
  • Sanitization: None detected.
  • [NO_CODE]: The skill consists entirely of markdown instruction files and rubrics. It does not include any executable scripts, binaries, or dependencies, which significantly limits the technical attack surface.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 15, 2026, 03:25 AM
Security Audit — agent-trust-hub — content-audit