playwright-recording

Pass

Audited by Gen Agent Trust Hub on Aug 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The reference documentation includes a utility function utilizing child_process.execSync to run the ffprobe command. This is used to extract the duration from recorded video files, with the execution restricted to local file paths generated by the tool.
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to download and install the playwright framework and required browser binaries from official software registries.
  • [DATA_EXFILTRATION]: The documentation describes patterns for persisting browser session state, including cookies and local storage, to a local auth.json file. While this is a standard feature for preserving authentication in automated workflows, the resulting file contains sensitive credentials.
  • [PROMPT_INJECTION]: The skill interacts with untrusted external content via automated browser navigation.
  • Ingestion points: Navigation to arbitrary external URLs via page.goto() in SKILL.md and reference.md.
  • Boundary markers: None; the skill does not explicitly instruct the agent to ignore instructions embedded in the target web pages.
  • Capability inventory: The skill possesses file system write access (fs.renameSync, fs.mkdirSync) and command execution capabilities (execSync).
  • Sanitization: The ffprobe utility function uses quoted paths for file arguments to mitigate potential command injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 1, 2026, 04:37 AM