cmd-interview-prep

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires the agent to read and write to the local file system within the directory /Users/olshansky/workspace/interviews/. This includes automated globbing of markdown files to build context and updating index files and reviews to persist session progress.
  • [EXTERNAL_DOWNLOADS]: The instructions include a reference to a Python cheatsheet hosted on Anthropic's Claude.ai platform.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it ingests content from a wide range of local markdown files to build its internal map of gaps and vocabulary holes. Evidence: Ingestion point is the recursive glob search of the workspace directory; boundary markers for separating file content from system instructions are absent; capabilities include writing to the local file system; and no explicit sanitization of the markdown file content is specified.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 03:22 PM