flyio-cli-public
Pass
Audited by Gen Agent Trust Hub on Mar 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
flyctlcommand-line tool to perform administrative tasks including application deployment, log streaming, and remote console access via SSH. - [COMMAND_EXECUTION]: Utilizes a vendor-provided script
scripts/fly_app_from_toml.shto extract application metadata from local configuration files. - [EXTERNAL_DOWNLOADS]: Recommends the use of official GitHub Actions provided by the service vendor for automated deployment and PR preview workflows.
- [PROMPT_INJECTION]: An indirect prompt injection surface is present when the agent processes untrusted data from commands such as
fly logsandfly status. Evidence: 1. Ingestion points:fly logsandfly status. 2. Boundary markers: None identified. 3. Capability inventory:fly deploy,fly ssh console,fly secrets set. 4. Sanitization: None identified. This risk is effectively mitigated by the skill's explicit requirement for user approval before executing any non-read-only commands.
Audit Metadata