jam-annotate
Fail
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill provides instructions to download and execute an installation script from https://native.jam.dev/install using the curl | bash pattern. This is a high-privilege operation used to install the Jam CLI required for the skill's functionality.\n- [EXTERNAL_DOWNLOADS]: The skill fetches the official Jam CLI installer from the vendor's own domain (native.jam.dev) as part of the setup process.\n- [COMMAND_EXECUTION]: The skill executes various shell commands to interact with the Jam service, including authentication checks (jam auth status), content creation (jam create jam), and feedback submission (jam feedback).\n- [INDIRECT_PROMPT_INJECTION]: The skill incorporates user-provided strings for fields like url, title, and description directly into the jam create jam command. This represents a potential command injection surface if the agent does not adequately escape these inputs.\n
- Ingestion points: User-provided text strings intended for the url, title, and description fields in the Jam payload.\n
- Boundary markers: None. The skill does not instruct the agent to use specific delimiters or ignore instructions within these user-supplied fields.\n
- Capability inventory: Shell subprocess execution (jam CLI) capable of network communication and file system interaction.\n
- Sanitization: No instructions are provided for the agent to sanitize, escape, or validate the user-controlled input before passing it to the shell command line.
Recommendations
- HIGH: Downloads and executes remote code from: https://native.jam.dev/install - DO NOT USE without thorough review
Audit Metadata