ironbee-node-devtools-cli

Warn

Audited by Gen Agent Trust Hub on May 29, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the ironbee-node-devtools-cli to interact with local system processes, manage daemons on local ports (2020, 2021), and connect to Docker containers for debugging purposes.
  • [REMOTE_CODE_EXECUTION]: The skill facilitates dynamic JavaScript execution through several mechanisms: the 'run execute' command allows running batch JavaScript within a session VM, and debug probes like tracepoints and logpoints support the runtime evaluation of conditions and expressions.
  • [EXTERNAL_DOWNLOADS]: The documentation instructs the user to install the '@ironbee-ai/devtools' package globally via npm and adds skills using npx. These resources are provided by the official skill author (ironbee-ai).
  • [PROMPT_INJECTION]: The skill exposes a surface for indirect prompt injection by ingesting data from external processes.
  • Ingestion points: Data enters the agent context through 'debug get-logs' and 'debug get-probe-snapshots' which capture console output and variable states from the debugged application.
  • Boundary markers: The CLI supports JSON output (--json) to structure the data, but it does not inherently prevent the agent from obeying instructions embedded in the logs or variable values.
  • Capability inventory: The agent can execute system commands and modify debugging state using the ironbee-node-devtools-cli tool.
  • Sanitization: No explicit sanitization or filtering of the captured application data is described.
Audit Metadata
Risk Level
MEDIUM
Analyzed
May 29, 2026, 01:38 AM
Security Audit — agent-trust-hub — ironbee-node-devtools-cli