code-walkthrough

Pass

Audited by Gen Agent Trust Hub on Jul 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to utilize termctrl to manage terminal sessions, execute Neovim, send interactive commands, and generate artifacts like recordings or screenshots. This provides the agent with high-level control over the local terminal environment.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests and interprets external source code and terminal output as part of the walkthrough process.
  • Ingestion points: Source files (e.g., .ts files) and the live terminal output from the termctrl session.
  • Boundary markers: Absent; there are no instructions to use delimiters or ignore potentially malicious commands embedded in code comments or program output.
  • Capability inventory: The skill can execute terminal commands via termctrl, save files to the local system, and manipulate the state of the active Neovim session.
  • Sanitization: Absent; the instructions do not specify any validation or filtering for the content read from files or the terminal before the agent processes it.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 19, 2026, 05:34 AM
Security Audit — agent-trust-hub — code-walkthrough