skills/kriscard/skills/weekly/Gen Agent Trust Hub

weekly

Pass

Audited by Gen Agent Trust Hub on Aug 13, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill reads external data from Obsidian daily notes and TIL files which represents an indirect injection surface. Ingestion points: Step 2 and Step 7 involve reading various local markdown files and project notes. Boundary markers: None identified in the instructions to prevent the agent from following instructions embedded in notes. Capability inventory: File system read and write operations via Obsidian tools and 'pwd' shell command execution. Sanitization: No explicit sanitization or filtering of note content is described.
  • [COMMAND_EXECUTION]: The skill executes the 'pwd' command in Step 5 to validate the environment and ensures the agent is in the correct working directory before proceeding with workspace-specific tasks.
  • [EXTERNAL_DOWNLOADS]: The skill references 'Obsidian CLI' as a dependency for its core note-management functionality.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 13, 2026, 03:59 PM
Security Audit — agent-trust-hub — weekly