tilt-dev
Pass
Audited by Gen Agent Trust Hub on Mar 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local commands using the
tiltCLI and provided shell scripts (check_tilt_status.sh,get_service_health.sh) to monitor and manage services.\n- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface where the agent processes potentially untrusted output from local service logs and health check responses.\n - Ingestion points: Service logs retrieved via
tilt logsand response bodies fromcurlhealth checks inscripts/get_service_health.sh.\n - Boundary markers: Absent; the skill does not use delimiters or instructions to notify the agent that service output should be treated as untrusted data.\n
- Capability inventory: The skill provides capabilities to execute
tiltcommands and shell scripts based on agent decisions.\n - Sanitization: Absent; no escaping or validation is performed on the output from local services before it is presented to the agent.
Audit Metadata