The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references the @robomotion/sdk Node.js package, which is the official SDK for the skill's authoring organization.
[DATA_EXFILTRATION]: While the skill involves capturing network traffic (including headers and bodies), it is a developer tool designed for this purpose. The instructions include a specific rule to "Store credentials in vault — never hardcode API keys or tokens," promoting secure data handling.
[DYNAMIC_EXECUTION]: The skill utilizes the Core.Programming.Function node to execute JavaScript logic for processing API responses. This is a standard feature of the Robomotion platform and the examples provided use static templates for data transformation.
[PROMPT_INJECTION]: The skill includes an attack surface for indirect prompt injection as it processes untrusted data from external network captures.
Ingestion points: Network request/response data is ingested via browser_get_request_response (documented in SKILL.md).
Boundary markers: No specific delimiters or "ignore" instructions are present in the provided templates to isolate external data from the agent's logic.
Capability inventory: The skill uses Core.Net.HttpRequest for network operations and Core.Programming.Function for code execution.
Sanitization: The provided code templates perform data extraction but do not demonstrate explicit sanitization or validation of the ingested network content.

reversing-network