cryptotokenkit

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill describes the process for registering macOS token extensions using the sudo -u _securityagent command within the token-and-smart-card-workflows.md reference. This is a legitimate administrative action required for system-level token integration.\n- [PRIVILEGE_ESCALATION]: Mentions the use of sudo for extension registration. This involves elevated privileges but is restricted to the specific deployment workflow described in the skill.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes structured data from external smart card hardware.\n
  • Ingestion points: APDU responses and TLV records retrieved from smart cards via TKSmartCard (cryptotokenkit-patterns.md).\n
  • Boundary markers: The instructions do not define specific delimiters for separating card data from agent prompts.\n
  • Capability inventory: Includes capabilities for data signing, decryption, and smart card PIN operations (token-and-smart-card-workflows.md).\n
  • Sanitization: Uses TKBERTLVRecord.sequenceOfRecords and TKSimpleTLVRecord for structured parsing of the hardware-provided data.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 11:13 AM
Security Audit — agent-trust-hub — cryptotokenkit