agent-cli-hint-completion

Pass

Audited by Gen Agent Trust Hub on Mar 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill facilitates the execution of various CLI tools (such as Playwright, Claude, and Gemini) through its own wrapper command 'agtm run'. This is a primary function of the skill for management and automation.
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the '@aiagenta2z/agtm' package from the NPM registry and fetches updated command hints and configurations from the vendor's GitHub repository.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing external data and executing commands based on that data.
  • Ingestion points: External data is ingested from URLs visited via integrated browser tools (like Playwright) and from user-supplied prompts for various third-party LLM CLIs.
  • Boundary markers: The documentation does not specify the use of delimiters or 'ignore' instructions to prevent the agent from obeying malicious commands embedded in processed external content.
  • Capability inventory: The skill possesses capabilities to execute subprocesses, perform network operations, and interact with the file system via the integrated tools it manages.
  • Sanitization: There is no description of data sanitization, validation, or filtering of the content retrieved from external sources before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 20, 2026, 10:05 AM
Security Audit — agent-trust-hub — agent-cli-hint-completion