Playwright Session Recorder

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns or security risks were identified in the skill's instructions or metadata. The skill utilizes well-known browser automation tools (Playwright) for its core functionality.
  • [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection because its primary function involves processing untrusted data from external web pages during browser recording sessions.
  • Ingestion points: Browser DOM snapshots and network HAR files captured from visited websites (SKILL.md).
  • Boundary markers: No explicit delimiters or boundary markers are defined to separate ingested web content from agent instructions.
  • Capability inventory: The skill records interactions and generates replayable scripts/logs based on external content (SKILL.md).
  • Sanitization: The skill documentation mentions "credential masking" during authentication flow capture, though the implementation details are not present in the markdown.
  • [EXTERNAL_DOWNLOADS]: The skill instructions include installation commands targeting the vendor's own repository and the well-known Playwright tool ecosystem.
  • [COMMAND_EXECUTION]: The skill description involves browser automation and script generation, which are standard functionalities for development-oriented agent skills.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 03:35 AM
Security Audit — agent-trust-hub — Playwright Session Recorder