The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external transaction data which could contain malicious instructions embedded in fields like transaction descriptions. This is an attack surface because the agent also has the capability to perform sensitive operations such as fund withdrawals.
Ingestion points: toobit account balance-flow (description field), toobit account audit, and toobit account info (token names).
Boundary markers: None identified; the skill does not instruct the agent to ignore or delimit instructions found within these data sources.
Capability inventory: account_withdraw (WRITE), account_sub_transfer (WRITE), and toobit account info (READ).
Sanitization: No explicit sanitization or validation of API response content is defined to prevent prompt injection.
[EXTERNAL_DOWNLOADS]: The skill requires the installation of the toobit-trade-cli package from npm, which is the vendor's tool for exchange interaction.
Package: toobit-trade-cli via npm.
Evidence: Installation is defined in the agent.install section and mentioned in the prerequisites.
[COMMAND_EXECUTION]: The skill executes the toobit command-line binary to perform account and portfolio operations.

toobit-cex-portfolio