truefoundry-service-test
Pass
Audited by Gen Agent Trust Hub on Apr 2, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides logic to automatically install the vendor's command-line interface and Python library from the standard package registry if they are not already present in the environment.
- [COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute shell commands for service health checks, smoke testing, and performance validation using tools like curl.
- [PROMPT_INJECTION]: The skill identifies a potential attack surface for indirect prompt injection within the responses of the services being tested. It incorporates specific guidance for the agent to treat these responses as untrusted third-party content and avoid interpreting instructions embedded in them.
- Ingestion points: HTTP response bodies returned from services during health and endpoint tests in
SKILL.md. - Boundary markers: Explicit security warnings in the Layer 3 instructions of the
SKILL.mdfile. - Capability inventory: Access to shell command execution through the
Bashtool and interaction with the TrueFoundry platform API. - Sanitization: Guidelines for the agent to parse only structured metadata and HTTP status codes rather than interpreting response body text.
Audit Metadata