anki-connect

Pass

Audited by Gen Agent Trust Hub on Aug 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on the execution of shell commands, specifically curl and jq, to communicate with the AnkiConnect API at http://127.0.0.1:8765. It provides templates for constructing JSON requests and parsing responses to manage decks, notes, and cards.
  • [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it reads and processes user-controlled data from Anki notes.
  • Ingestion points: Data retrieved from Anki via actions like notesInfo, cardsInfo, and findNotes as described in SKILL.md.
  • Boundary markers: The skill does not define specific delimiters or instructions to prevent the agent from obeying instructions embedded within the flashcard content.
  • Capability inventory: The agent can perform network requests to the local API and modify or delete local flashcard data.
  • Sanitization: While the skill provides patterns for checking API errors, it does not include instructions for sanitizing or escaping the content retrieved from the notes themselves.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 19, 2026, 05:43 AM
Security Audit — agent-trust-hub — anki-connect