ln-700-project-bootstrap

Pass

Audited by Gen Agent Trust Hub on Apr 25, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill utilizes a fallback mechanism to fetch shared files and reference documentation from the author's official GitHub repository (github.com/levnikolaevich/claude-code-skills) using the WebFetch tool when local directories are absent.\n- [COMMAND_EXECUTION]: The bootstrap workflow and verification procedures involve executing standard development CLI tools including npm, dotnet, docker-compose, and curl to manage dependencies, build projects, and verify service health.\n- [PROMPT_INJECTION]: During the technology detection phase, the skill ingests metadata from local project files (e.g., package.json, .csproj, requirements.txt), which constitutes a potential surface for indirect prompt injection from untrusted codebases.\n
  • Ingestion points: Detection logic in SKILL.md and references/stack_detection.md scans the project root and subdirectories.\n
  • Boundary markers: No specific delimiters or "ignore instructions" warnings are applied to the ingested file content before it is processed by the agent.\n
  • Capability inventory: The skill and its delegated sub-skills possess extensive system capabilities, including file creation, script execution, and container management.\n
  • Sanitization: Ingested file contents are parsed for technology patterns and versions but are not explicitly sanitized or escaped.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 25, 2026, 07:29 PM