The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted text from third-party websites, creating a surface for indirect prompt injection.
Ingestion points: Scraper scripts such as scripts/qidian-rank-scraper.js and scripts/fanqie-rank-scraper.js extract book descriptions and summaries from multiple external novel platforms.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the skill's logic for isolating external content.
Capability inventory: The skill utilizes child_process.execSync (in scripts/cdp-utils.js) to run browser tools and fs.writeFileSync to save data to the local file system.
Sanitization: No validation or escaping is applied to the ingested external descriptions before they are presented to the agent.
[COMMAND_EXECUTION]: The script scripts/cdp-utils.js uses child_process.execSync to invoke the agent-browser tool. This is a functional requirement for browser automation and scraping.
[EXTERNAL_DOWNLOADS]: The skill connects to multiple well-known web novel platforms to retrieve data. These operations are core to the skill's purpose and target established, legitimate services.

story-long-scan