The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the host to add an external Model Context Protocol (MCP) server located at https://api.wellapp.ai/v1/mcp. This server provides the tools (well_*) necessary for the skill to function.
[DATA_EXFILTRATION]: The skill accesses and processes sensitive financial data including invoices, transactions, and workspace details. This information is exchanged with the vendor's API endpoint. As the data flow is essential for the reconciliation purpose and targets the vendor's own infrastructure, this is considered functional behavior.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it retrieves and processes external financial records which could contain malicious instructions.
Ingestion points: Data is ingested via well_query_records and well_get_schema in the SKILL.md workflow.
Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potentially malicious content within the retrieved records.
Capability inventory: The skill utilizes tools to query sensitive databases (invoices, transactions, invoice_transactions).
Sanitization: There is no evidence of sanitization or filtering of the retrieved record content before it is processed by the agent.

payment-invoice-lookup