The Agent Skills Directory

[PROMPT_INJECTION]: No malicious patterns or attempts to override agent behavior were identified. The instructions use standard, benign language to define the skill's operational boundaries.- [DATA_EXFILTRATION]: The skill handles trip data locally within the trips/logs/ directory. It explicitly mandates that the agent should not store .env values or account numbers in these logs, adhering to data privacy best practices. Furthermore, it notes that the storage directory should be gitignored to prevent accidental public disclosure.- [COMMAND_EXECUTION]: The skill utilizes standard grep commands to facilitate the retrieval of past trip logs. These operations are limited to the specific trips/logs/ directory and do not involve administrative privileges or piping data to external shells.- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external flight searches and existing markdown logs, which creates a theoretical attack surface for indirect injection if that data were maliciously crafted.
Ingestion points: Reading content from the trips/logs/ directory via grep.
Boundary markers: Absent; there are no specific instructions to ignore embedded commands within the log files.
Capability inventory: File system write access and grep execution capabilities.
Sanitization: The skill does not define specific sanitization routines, relying on the agent's default handling of text data.

trip-log