The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were detected. The skill follows best practices for tool documentation by providing a lightweight stub that loads content dynamically.
[COMMAND_EXECUTION]: The skill executes the agent-browser CLI via the Bash tool. This command execution is the primary intended function for performing browser automation and retrieving documentation.
[EXTERNAL_DOWNLOADS]: Additional instructions and specialized skills are fetched from the official CLI infrastructure provided by the developer. These resources originate from a well-known and trusted organization and are necessary for the skill's functionality.
[PROMPT_INJECTION]: The skill design creates a surface for indirect prompt injection by incorporating external data into the agent's context.
Ingestion points: External content enters the agent's context through the output of agent-browser skills get commands as specified in SKILL.md.
Boundary markers: No specific delimiters or boundary instructions are present to differentiate the fetched content from system-level instructions.
Capability inventory: The skill allows execution of agent-browser via Bash for website interaction and automation.
Sanitization: Content retrieved from the CLI is provided to the agent without additional sanitization or filtering layers.

agent-browser