The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill utilizes npx to download and run the @useclawlink/cli package from the official NPM registry. This is a common and expected method for utilizing the integration's command-line tools.
[COMMAND_EXECUTION]: The instructions direct the agent to execute shell commands via the @useclawlink/cli tool to manage Zoho Books data. This includes logging in, connecting the account, and running specific accounting actions.
[DATA_EXFILTRATION]: Local storage of authentication credentials at ~/.clawlink/credentials.json is used to maintain the session. This follows standard CLI application practices for credential management.
[INDIRECT_PROMPT_INJECTION]: The skill interacts with external data (invoices, contacts, bills) from Zoho Books, which could potentially contain instructions that influence the agent's behavior.
Ingestion points: Data enters the agent context through actions such as zoho_books_list_invoices and zoho_books_get_contact executed via the CLI in SKILL.md.
Boundary markers: There are no explicit markers or instructions used to isolate retrieved data from the agent's core instructions.
Capability inventory: The skill possesses the capability to execute shell commands through the npx environment.
Sanitization: No explicit sanitization or validation of the data retrieved from Zoho Books is mentioned in the skill instructions.

zoho-books