The Agent Skills Directory

[SAFE]: The skill uses the Playwright MCP plugin to interact with Bing.com, a well-known and trusted search engine. All browser operations (navigation, typing, and script evaluation) are consistent with the stated purpose of performing web searches and extracting results.\n- [SAFE]: The JavaScript scripts included with the skill ('scripts/extract-results.js' and 'scripts/search.js') are focused on standard DOM manipulation and data extraction. The use of 'atob' in the extraction script is a functional requirement for decoding Bing's internal URL redirect format and does not represent malicious obfuscation.\n- [SAFE]: The skill does not access sensitive local files, environment variables, or hardcoded credentials. It does not perform unauthorized network requests to untrusted domains.\n- [SAFE]: The skill ingests untrusted data from Bing search results. 1. Ingestion points: 'scripts/extract-results.js' extracts titles and descriptions from the Bing results page. 2. Boundary markers: 'SKILL.md' defines a markdown structure for the agent to follow when returning results. 3. Capability inventory: 'SKILL.md' utilizes browser navigation, typing, and script evaluation tools. 4. Sanitization: No sanitization is applied to the extracted web content. This surface is expected for a search utility and does not represent a malicious finding.

bing-search