learn-agent-skills
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches curriculum manifests, lesson documentation, and quiz data from the author's GitHub repository at fancyboi999/ai-engineering-from-scratch-zh.
- [COMMAND_EXECUTION]: Executes shell commands to check the presence and versions of Node.js, npx, and Python 3 on the learner's local system to ensure lab readiness.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted Markdown and JSON data from external URLs which could potentially contain malicious instructions intended to influence the agent's behavior.
- Ingestion points: Remote content is fetched from raw.githubusercontent.com for curriculum progression.
- Boundary markers: No explicit instruction delimiters or warnings to ignore embedded commands are present in the processing logic.
- Capability inventory: The skill has the capability to execute shell commands and write to the local file system (AGENT-SKILLS-LEARNING.md).
- Sanitization: There is no evidence of sanitization, escaping, or schema validation of the external content before it is interpolated into the agent's context.
Audit Metadata