anki-vocab

Pass

Audited by Gen Agent Trust Hub on Jul 12, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill operates entirely on local text files to manage vocabulary learning. It reads contextual information from files like mission.md and vocabulary.md to generate Anki-compatible text files.\n- [PROMPT_INJECTION]: The skill has an ingestion surface for indirect prompt injection as it reads from external files (mission.md, path.md, vocabulary.md, review-queue.md) that may contain untrusted data.\n
  • Ingestion points: Content is pulled from mission.md, path.md, vocabulary.md, and review-queue.md to construct sentences.\n
  • Boundary markers: None identified in the prompt instructions to distinguish between skill instructions and ingested data content.\n
  • Capability inventory: Limited to writing plain text files (.txt) in the anki/ directory and appending entries to markdown files (.md). No network access or shell execution capabilities are present.\n
  • Sanitization: The skill does not specify explicit sanitization or filtering of the ingested content, though it does avoid using specific separators like semicolons within fields.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 12, 2026, 05:26 PM
Security Audit — agent-trust-hub — anki-vocab