The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill utilizes the WebFetch tool to retrieve book metadata from well-known bibliographic services, including Douban and Dangdang. These operations are strictly limited to the skill's stated purpose of book research.
[PROMPT_INJECTION]: The skill processes data from external web sources, which represents a potential surface for indirect prompt injection.
Ingestion points: Content retrieved from book database sites via WebFetch and user-provided book titles.
Boundary markers: Sub-agent prompts use task templates but do not employ specific delimiters to isolate external content from instructions.
Capability inventory: The skill has access to WebSearch, WebFetch, Agent, and Write tools.
Sanitization: No explicit sanitization of web content is performed before processing by the sub-agents.
[DATA_EXFILTRATION]: No hardcoded credentials, API keys, or unauthorized network exfiltration of sensitive local data were detected. The skill only writes resulting reports to a local file.
[SAFE]: Analysis of the instruction set found no evidence of base64 obfuscation, hidden characters, persistence mechanisms, or attempts to acquire elevated system privileges.

book-analysis