skills/azhi-ss/ljg-skills/ljg-writes/Gen Agent Trust Hub

ljg-writes

Pass

Audited by Gen Agent Trust Hub on May 12, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to use the system date command to generate formatted timestamps and identifiers for file metadata.
  • Evidence: Found in sections 'Denote 文件规范' and 'Org 文件头', specifying commands like date +%Y%m%dT%H%M%S and date "+%Y-%m-%d %a %H:%M".
  • [DATA_EXFILTRATION]: The skill defines a specific local path for saving output files, which involves file system write operations.
  • Evidence: Section 'Denote 文件规范' identifies ~/Documents/notes/ as the designated output directory for generated .org files.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it processes user-provided viewpoints to generate articles.
  • Ingestion points: User input provided in the 'Process' section under '把观点放到台面上'.
  • Boundary markers: Absent; there are no explicit delimiters or instructions to ignore embedded commands within the user viewpoint.
  • Capability inventory: The skill has the ability to execute shell commands (date) and write to the local file system (~/Documents/notes/).
  • Sanitization: Absent; the skill does not specify any filtering or validation for the input viewpoint. However, the strict enforcement of Org-mode syntax and ASCII-only diagrams significantly limits the potential for malicious instruction execution.
Audit Metadata
Risk Level
SAFE
Analyzed
May 12, 2026, 07:13 PM