output-scoring-en

Pass

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill defines a procedural rubric for scoring legal text and does not include any executable scripts or binary files.
  • [EXTERNAL_DOWNLOADS]: The documentation references a public research repository from Fudan University for attribution; it does not download or execute code from this source.
  • [PROMPT_INJECTION]: The skill identifies a potential surface for indirect prompt injection as it processes untrusted AI legal outputs.
    1. Ingestion points: AI legal output processed for scoring in SKILL.md.
    2. Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present.
    3. Capability inventory: Restricted to the 'Read' tool; no execution or network tools allowed.
    4. Sanitization: No content filtering or sanitization is specified. Analysis: Due to the lack of high-privilege tools, the risk of exploitation is negligible.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 14, 2026, 08:28 AM
Security Audit — agent-trust-hub — output-scoring-en