The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes external booking data from the Navan REST API (api.navan.com). While the data is processed using structured tools like jq and TypeScript interfaces, the processing of untrusted external content represents a potential surface for indirect prompt injection.
Ingestion points: Data fetched from https://api.navan.com/v1/bookings and saved to bookings.json (SKILL.md).
Boundary markers: None explicitly implemented; the skill assumes the structure of the API response.
Capability inventory: The skill uses Bash(curl), Write, and Edit tools to interact with the system and files.
Sanitization: No explicit sanitization or filtering of API response text fields is performed prior to processing.
[DATA_EXPOSURE]: The skill correctly instructs users to manage sensitive credentials (NAVAN_CLIENT_ID, NAVAN_CLIENT_SECRET) using environment variables, which is a standard and safe practice for secret management.
[COMMAND_EXECUTION]: The skill utilizes standard command-line tools (curl, jq) via the Bash tool to interact with well-known Navan API endpoints. The commands are scoped to the skill's primary purpose of travel spend analytics and optimization.

navan-cost-tuning