The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes standard command-line utilities such as curl, jq, and grep to perform API requests and process response data. These operations are explicitly allowed by the skill's metadata and are consistent with its stated purpose of API management.\n- [DATA_EXFILTRATION]: Network activity is restricted to the official Navan API domain (api.navan.com). Authentication is handled securely using environment variables (NAVAN_CLIENT_ID, NAVAN_CLIENT_SECRET), ensuring that sensitive credentials are not hardcoded or exposed within the skill code.\n- [INDIRECT_PROMPT_INJECTION]:\n
Ingestion points: The skill retrieves external data from https://api.navan.com/v1/ as shown in the instructions in SKILL.md.\n
Boundary markers: There are no explicit boundary markers or instructions for the agent to ignore embedded commands within the fetched data.\n
Capability inventory: The skill has access to Bash, Write, and Edit tools which could be influenced by external input.\n
Sanitization: The skill uses jq to parse and validate the structure of API responses, which acts as an implicit structural filter, although no explicit content-level sanitization is performed.

navan-upgrade-migration