The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the retrieval of external content from the QQ platform (e.g., user profiles, member lists), which serves as an entry point for untrusted data that could contain malicious instructions.
Ingestion points: Tools qq_list_members, qq_member_detail, qq_list_guilds, and qq_list_channels as referenced in SKILL.md.
Boundary markers: Absent. The instructions do not define delimiters or specific warnings to ignore instructions found within retrieved data.
Capability inventory: High-privilege administrative tools such as qq_kick_member, qq_mute_member, and qq_add_role listed in SKILL.md.
Sanitization: Absent. There is no evidence of input validation or content filtering for the data fetched from the QQ API.
[NO_CODE]: The skill does not contain any executable scripts, dependencies, or compiled binaries, relying entirely on provided markdown instructions and platform tools.

qq