retirement-optimizer

Warn

Audited by Gen Agent Trust Hub on Mar 23, 2026

Risk Level: MEDIUMPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [DATA_EXFILTRATION]: The skill accesses the home directory path ~/.claude/projects/ to append execution metadata to skill-telemetry.md. While used for tracking, access to hidden platform-specific directories in the user home is a sensitive operation.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted data from external codebases and data models without sufficient isolation.
  • Ingestion points: Processes package manifests, core data structures (user profiles, account balances), and external data sources (market data, tax brackets) across multiple analysis phases.
  • Boundary markers: Absent. The instructions do not define clear delimiters or include 'ignore' warnings for the data being analyzed.
  • Capability inventory: The skill has file-write capabilities, creating reports in docs/ and modifying telemetry files in ~/.claude/projects/.
  • Sanitization: Absent. There is no evidence of validation or escaping logic for the codebase content or data models before they are processed by the agent.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 23, 2026, 10:57 AM
Security Audit — agent-trust-hub — retirement-optimizer