gcse-english-language-tutor
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: No malicious instructions or attempts to bypass safety guardrails were found. The skill maintains a consistent educational persona and strictly adheres to its tutoring purpose and the AQA 2026 specification constraints.\n- [DATA_EXFILTRATION]: No evidence of data collection or exfiltration was detected. The skill uses local markdown references in the references/ directory and does not include network commands (like curl or wget) or hardcoded credentials.\n- [OBFUSCATION]: The skill uses clear, human-readable markdown. No hidden text, Base64 encoding, zero-width characters, or homoglyph-based URL manipulation was found across the skill or its reference files.\n- [REMOTE_CODE_EXECUTION]: The skill does not download or execute external scripts or packages. It relies entirely on static instructional content and local references.\n- [INDIRECT_PROMPT_INJECTION]: The skill interacts with user-supplied text for feedback, which constitutes an ingestion point for untrusted data. However, the skill has no functional capabilities that could be leveraged if a prompt injection occurred.\n
- Ingestion points: User-provided essays and queries in the chat interface.\n
- Boundary markers: Not explicitly defined in the tutor instructions.\n
- Capability inventory: None detected; the skill is restricted to text interaction.\n
- Sanitization: None described, as the skill lacks code for processing strings beyond simple text display.\n- [PRIVILEGE_ESCALATION]: No commands related to administrative privileges or system modification (such as sudo or chmod) were detected.
Audit Metadata