app-store-screenshots
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: CRITICALREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [REMOTE_CODE_EXECUTION] (CRITICAL): Untrusted remote script execution via shell pipe.
- Evidence: The skill contains the command
curl -fsSL https://cli.inference.sh | shin the 'Quick Start' section. - Risk: This pattern downloads a script from a non-whitelisted domain and executes it directly in the system shell without verification. This allows the domain owner to execute arbitrary code on the host machine.
- [COMMAND_EXECUTION] (HIGH): Execution of third-party CLI tools with broad permissions.
- Evidence: The skill extensively uses the
infshtool, which is defined in the frontmatterallowed-tools: Bash(infsh *)and used for operations likeinfsh loginandinfsh app run. - Risk: Since the installation of this tool is untrusted, all subsequent calls to the tool inherit the same risk of arbitrary code execution.
- [EXTERNAL_DOWNLOADS] (MEDIUM): Unverifiable external dependencies.
- Evidence: The 'Related Skills' section encourages installing additional skills via
npx skills add inferencesh/skills@.... - Risk: This expands the attack surface by pulling in more external, unvetted code from the same untrusted source.
Recommendations
- CRITICAL: Downloads and executes remote code from untrusted source(s): https://cli.inference.sh - DO NOT USE
- AI detected serious security threats
Audit Metadata