skills/johnnydecimal/skills/jdex/Gen Agent Trust Hub

jdex

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches system configurations and documentation from the official vendor MCP server (https://johnnydecimal.com/mcp).
  • [REMOTE_CODE_EXECUTION]: Provides instructions to call an install_cli function on the remote server to download and set up the jd command-line tool locally.
  • [COMMAND_EXECUTION]: Employs shell utilities (ls, find) and local CLI tools (jd, obsidian) to perform filesystem indexing, note management, and file relocation tasks.
  • [INDIRECT_PROMPT_INJECTION]: Accesses and processes user-controlled markdown notes and file metadata, creating a surface for indirect instructions to influence agent behavior.
  • Ingestion points: Johnny.Decimal index notes (JDex), file metadata, and file content.
  • Boundary markers: Relies on metadata separators but lacks explicit instructions to sanitize or ignore embedded prompts in user files.
  • Capability inventory: File modification, directory traversal, and execution of local management tools.
  • Sanitization: No specific content filtering or sanitization logic is provided for ingested data.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 01:00 PM
Security Audit — agent-trust-hub — jdex