Skills
skills/gasserane/personal-skills/learning-agenda/Gen Agent Trust Hub

learning-agenda

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for Indirect Prompt Injection by ingesting and processing untrusted external data.
  • Ingestion points: User-provided inputs defined in SKILL.md, specifically Question list (input 4), Status updates (input 7), MA strategic context (input 9), and MA question list (input 10).
  • Boundary markers: The instructions do not define clear delimiters or 'ignore embedded instructions' warnings for these user-supplied inputs.
  • Capability inventory: The skill uses internal tools for file operations, including reading files from agent-improvements/ and mel_wiki/, and writing reports using ane_package.reporting.word_export.write_word_report and ane_package.reporting.pdf_export.write_pdf_report.
  • Sanitization: There is no mention of input validation, escaping, or filtering for the user-supplied text before it is interpolated into the final document structure.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 08:50 AM
Security Audit — agent-trust-hub — learning-agenda