Skills
Navigate
skills/cskwork/agent-skills/skill-usage-stats/Gen Agent Trust Hub

skill-usage-stats

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The analyze.py script uses the Python webbrowser module to automatically open the generated HTML report in the user's default browser after scanning is complete.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted interaction data from Claude Code and Codex session transcripts, which creates an indirect prompt injection surface.
  • Ingestion points: The analyze.py script reads .jsonl interaction logs from the user's home directory at ~/.claude/projects/ and ~/.codex/sessions/.
  • Boundary markers: None identified in the transcript parsing logic; the script treats all transcript lines as potential data sources.
  • Capability inventory: The skill has the capability to write files to the local disk (~/.claude/skill-stats.html) and launch the default web browser.
  • Sanitization: The generated HTML report implements a JavaScript esc() function that escapes <, >, and & characters before rendering transcript-derived data into the dashboard, providing a layer of protection against cross-site scripting (XSS) from malicious content within the logs.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 10:33 AM