international-expansion

Pass

Audited by Gen Agent Trust Hub on Jul 3, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONNO_CODE
Full Analysis
  • [NO_CODE]: The skill consists exclusively of instructional Markdown and YAML configuration. It does not include or execute external scripts, binary files, or library dependencies.
  • [SAFE]: The skill reads matter-specific context and jurisdictional footprints from ~/.claude/plugins/config/claude-for-legal/employment-legal/CLAUDE.md. This file access is limited to the skill's specific application directory and follows legitimate plugin logic for contextual awareness.
  • [COMMAND_EXECUTION]: The skill generates and writes persistent tracker data to a YAML file at ~/.claude/plugins/config/claude-for-legal/employment-legal/expansion-[country-slug].yaml. This file-writing behavior is a functional component of the workflow and is restricted to the plugin's local directory.
  • [PROMPT_INJECTION]: The skill identified a surface for Indirect Prompt Injection during the processing of user data.
  • Ingestion points: Information gathering step (SKILL.md) where users provide details on roles and country context.
  • Boundary markers: None present to delimit user-provided data within the generated reports or YAML files, potentially allowing embedded instructions to be processed by downstream tools.
  • Capability inventory: File system write access to specific configuration paths and interpolation of data into briefing requests.
  • Sanitization: The skill specifies a lowercase-hyphenated format for the filename slug to prevent directory traversal, but lacks explicit sanitization or escaping for the contents of the generated tracker items.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 3, 2026, 03:58 PM
Security Audit — agent-trust-hub — international-expansion