The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external websites using Playwright's data extraction methods.
Ingestion points: The skill uses page.evaluate(), page.content(), and page.title() (found in SKILL.md) to extract text and HTML from remote web pages.
Boundary markers: No boundary markers or "ignore embedded instructions" warnings are implemented in the provided patterns.
Capability inventory: The skill has access to the Bash tool to execute shell commands and the Write tool to create files, creating a potential path for code execution or file system modification if malicious instructions are processed.
Sanitization: There is no evidence of sanitization or filtering of the extracted web content before it is processed by the agent.
[COMMAND_EXECUTION]: The skill relies on shell commands for environment setup and process control.
Evidence includes using pkill to terminate existing Chromium instances, launching chromium with remote debugging flags, and running node to execute generated automation scripts.
[DYNAMIC_EXECUTION]: The skill follows a pattern of generating executable content at runtime.
The documentation instructs the agent to write JavaScript logic to .mjs files (via the Write tool) and then execute those files using node.
It also utilizes page.evaluate() to execute arbitrary JavaScript within the browser's context.

playwright