tutorial-writer

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill's core workflow requires the agent to execute arbitrary shell commands and run generated scripts (Node.js, Python, etc.) to verify tutorial accuracy. Specifically, the 'Testing' phase (Step 5 in SKILL.md and throughout TESTING-VERIFICATION.md) instructs the agent to create isolated environments, install packages via npm/pip, and execute code examples start-to-finish without skipping steps.
  • [PROMPT_INJECTION]: The skill exhibits a vulnerability surface for indirect prompt injection. It ingests external, potentially untrusted source code to generate tutorial content and then immediately executes commands derived from that content during the verification phase.
  • Ingestion points: The agent reads source files for APIs (Step 2 in SKILL.md) and test files for usage patterns (Step 1.3 in TESTING-VERIFICATION.md).
  • Boundary markers: The instructions do not specify the use of delimiters or warnings to ignore embedded instructions within the source files being analyzed.
  • Capability inventory: The skill uses subprocess execution (bash, node, python), file system write operations (creating test directories and script files), and network operations (package installation via npm/pip).
  • Sanitization: There is no mention of sanitizing or validating the commands extracted from the source code or tutorial steps before execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 07:48 AM