Skills
Navigate
skills/yennanliu/cs_basics/lc-site-data/Gen Agent Trust Hub

lc-site-data

Pass

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Executes local project scripts including site/build.sh, script/fetch_problem_lists.py, and npm test to validate data and build site components. It also utilizes python3 -m http.server to host a local preview of the site.
  • [DYNAMIC_EXECUTION]: Employs inline Python commands (python3 -c) to parse JSON data and extract specific identifiers during the verification phase.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from README.md and JSON files that could theoretically contain malicious instructions if the source repository content is untrusted.
  • Ingestion points: Reads data from data/roadmap.json, data/complexity_quiz.json, and the project README.md.
  • Boundary markers: The skill does not define explicit delimiters for ingested data but relies on build-time validation scripts to catch structural errors.
  • Capability inventory: Access to Bash (shell execution), Write, and Edit (modifying local project files).
  • Sanitization: Employs site/build-roadmap.js and site/build-quiz.js as strict validators that enforce data constraints and schema compliance.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 9, 2026, 08:54 AM