mcpa-certification

Pass

Audited by Gen Agent Trust Hub on Oct 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses python3 to run practical labs (main.py) and execute unit tests (unittest) found within the repository structure.
  • [DYNAMIC_EXECUTION]: Script execution is dynamic, as the agent is instructed to locate and run Python files based on paths defined in external configuration files like mcpa-f.json.
  • [EXTERNAL_DOWNLOADS]: The skill fetches certification program data, lesson Markdown, quizzes, and scripts from the rohitg00/ai-engineering-from-scratch repository on GitHub when a local clone is unavailable.
  • [INDIRECT_PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection due to its ingestion of external content.
  • Ingestion points: Reads data from program.json, mcpa-f.json, lesson Markdown files (en.md), and quiz.json (found in SKILL.md).
  • Boundary markers: The instructions do not define explicit boundary markers or warnings for the agent to ignore instructions embedded within the curriculum data.
  • Capability inventory: The skill has the capability to execute shell commands via python3 (found in SKILL.md).
  • Sanitization: There is no evidence of sanitization or filtering applied to the fetched educational content before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 30, 2026, 01:38 AM