rt-desk-reject-risk

Pass

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [NO_CODE]: The skill is entirely defined within a Markdown file and does not include any scripts or binaries.
  • [PROMPT_INJECTION]: The manuscript audit process introduces a surface for indirect prompt injection. While an adversary could include instructions in a manuscript to manipulate the audit results, the risk is negligible due to the absence of sensitive capabilities.
  • Ingestion points: User-provided manuscripts.
  • Boundary markers: The manuscript is treated as the analysis target, but no explicit delimiters are specified in the instructions.
  • Capability inventory: The skill is limited to reading local repository files (venue profiles) and performing text analysis.
  • Sanitization: No sanitization is performed on input text to filter out adversarial prompts.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 24, 2026, 10:30 AM
Security Audit — agent-trust-hub — rt-desk-reject-risk