tcm-constitution-analyzer

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill requests access to sensitive local files containing personal information and health records.
  • Evidence: The skill reads data from data-example/profile.json, data-example/nutrition-tracker.json, data-example/fitness-tracker.json, and data-example/sleep-tracker.json.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from multiple external health-tracking JSON files which could serve as an attack vector if the source files are compromised.
  • Ingestion points: Data is ingested from multiple JSON files in the data-example/ directory, including daily constitution logs.
  • Boundary markers: The instructions lack explicit delimiters or "ignore embedded instructions" warnings to isolate processed data from the agent's control flow.
  • Capability inventory: The skill has access to Read, Grep, Glob, and Write tools, allowing it to both access sensitive data and modify files on the system.
  • Sanitization: No input validation, escaping, or filtering mechanisms are described for the content retrieved from external tracker files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 07:58 AM
Security Audit — agent-trust-hub — tcm-constitution-analyzer