leave-tracker

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • Indirect Prompt Injection Surface: The skill ingests data from external sources, including HRIS integrations, local YAML configuration files, and user-uploaded spreadsheets. This creates a surface where the agent might process instructions embedded within that data.\n
  • Ingestion points: Data enters the agent context from ~/.claude/plugins/config/claude-for-legal/employment-legal/leave-register.yaml, connected HRIS systems, and attorney-uploaded leave spreadsheets (SKILL.md).\n
  • Boundary markers: The instructions do not specify the use of clear delimiters or instructions to ignore embedded prompts within the ingested leave data.\n
  • Capability inventory: The skill analyzes leave records to identify deadlines and explain required actions, involving summarization and natural language generation based on the input data.\n
  • Sanitization: No explicit sanitization or schema validation processes for the ingested spreadsheets or YAML data are described in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 07:01 PM
Security Audit — agent-trust-hub — leave-tracker