The Agent Skills Directory

[PRIVILEGE_ESCALATION]: The skill instructs the user to execute commands with elevated privileges using sudo to register macOS token extensions.
Evidence: sudo -u _securityagent /Applications/TokenHost.app/Contents/MacOS/TokenHost in references/token-and-smart-card-workflows.md.
[COMMAND_EXECUTION]: Provides shell commands for environment configuration and extension registration.
Evidence: Registration command in evals/evals.json and references/token-and-smart-card-workflows.md.
[EXTERNAL_DOWNLOADS]: Includes links to external documentation hosted on sosumi.ai, which is not a recognized trusted vendor domain.
Evidence: Multiple links in SKILL.md and references/token-and-smart-card-workflows.md.
[INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of data from external smart-card hardware via APDU responses and TLV parsing.
Ingestion points: TKSmartCard.send and TKSmartCard.transmit in references/cryptotokenkit-patterns.md.
Boundary markers: The skill emphasizes checking status words (e.g., 0x9000) but does not provide specific delimiters for separating hardware data from the LLM prompt context.
Capability inventory: The skill utilizes system-level APIs for token management and privilege escalation for registration.
Sanitization: Provides TLV parsing logic (TKBERTLVRecord), but does not include explicit sanitization of the values before they might be processed by an agent.

cryptotokenkit