tax-computation

Pass

Audited by Gen Agent Trust Hub on Jul 9, 2026

Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface.
  • Ingestion points: The skill reads several external files including firm-profile.md, plugin configurations (CLAUDE.md), references/tax_malaysia.md, and active client engagement documents (SKILL.md).
  • Boundary markers: No explicit delimiters or instructions are provided to the agent to treat loaded content as data or to ignore embedded instructions within those files.
  • Capability inventory: The skill performs file writes to create tax computation schedules and accounting journals on the local disk (SKILL.md).
  • Sanitization: The instructions do not specify any validation or sanitization of the content loaded from external files before it is processed or written back to disk.
  • [DATA_EXFILTRATION]: Local File Access.
  • The skill is designed to read files from the user's home directory (~/.claude/plugins/config/claude-for-accounting/firm-profile.md and ~/.claude/plugins/config/claude-for-accounting/{{plugin}}/CLAUDE.md). Accessing files within the home directory (~/) represents a potential data exposure risk, although in this context, the files appear to be configuration data specific to the accounting plugin.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 9, 2026, 02:12 PM
Security Audit — agent-trust-hub — tax-computation