supermemory-status

Warn

Audited by Gen Agent Trust Hub on Nov 17, 2026

Risk Level: MEDIUMDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The script scripts/status.js uses require() with a computed path involving directory traversal (path.join(__dirname, '..', '..', '..', 'scripts', 'cli.js')). This technique allows the skill to load and execute code located three levels above its own directory, which bypasses encapsulation and relies on a specific external file structure.
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute a bundled Node.js script using the node command. The execution includes the --cwd argument, which allows the script to operate within the user's workspace root.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes output from a local command and instructs the agent to interpret these results to inform its response to the user.
  • Ingestion points: Terminal output generated by the status.js script.
  • Boundary markers: None; the instructions do not specify delimiters or warnings for the agent to ignore potentially malicious content in the status output.
  • Capability inventory: Execution of Node.js scripts with access to the file system (workspace root) and network (API reachability checks).
  • Sanitization: There is no evidence of sanitization or validation performed on the output of the status probe before it is processed by the agent.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Nov 17, 2026, 01:35 PM