The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from co-ownership definition files and financial transaction data retrieved via the Qonto API. Ingestion points: The agent reads data from JSON files in the copros/ directory and transaction records in data/transactions/ into its active context. Boundary markers: The instructions do not define explicit delimiters or 'ignore' directives to prevent the agent from potentially obeying instructions embedded within co-ownership descriptions or transaction notes. Capability inventory: The skill performs file system operations (updating management journals) and instructs the agent to execute local scripts (node integrations/qonto/fetch.js). Sanitization: The instructions do not prescribe validation or sanitization logic for the content of the processed data files, relying on the underlying model's safety guardrails.

syndic