terra-routes
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to retrieve and process an external markdown file (sport-specific-examples.md) from the vendor's documentation portal. This creates a vulnerability surface where the agent's behavior could be influenced by content within the remote file.
- Ingestion point: The SKILL.md file directs the agent to fetch and read content from https://docs.tryterra.co/routes-api-pre-release/sport-specific-examples.md.
- Boundary markers: The skill does not provide delimiters or instructions to treat the fetched content as data rather than instructions.
- Capability inventory: The skill describes the use of network tools (curl) and a vendor CLI (@tryterra/cli) to manage remote API resources.
- Sanitization: There is no requirement for the agent to sanitize or validate the content of the fetched file before using it to construct API payloads.
- [EXTERNAL_DOWNLOADS]: The skill references live documentation and code examples from the tryterra.co domain. It also suggests installing the @tryterra/cli Node.js package and the tryterra/tap/terra Homebrew tap, which are the official tools provided by the skill author.
- [COMMAND_EXECUTION]: The skill provides templates for shell commands using curl to interact with the Terra API. It also recommends using the terra CLI tool to list references and manage integrations.
Audit Metadata