jam-to-e2e
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill documentation provides instructions to set up the Jam CLI by downloading and executing a script from the developer's domain (native.jam.dev) using a shell pipe.
- [COMMAND_EXECUTION]: The skill utilizes several shell commands and project-specific tools, including npm init, npx, and the jam CLI, to manage test runners and process recording metadata.
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources, such as video transcripts and user events from Jam recordings, to generate test cases and code.
- Ingestion points: External data is fetched through tools like getVideoTranscript, getDetails, and getUserEvents.
- Boundary markers: The instructions do not include specific delimiters or warnings to prevent the agent from following instructions that may be embedded within the Jam recordings.
- Capability inventory: The agent can write to the local file system and execute shell commands to install dependencies and run tests.
- Sanitization: The skill lacks explicit sanitization steps for the data retrieved from external recordings before it is used to influence code generation.
Audit Metadata