The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it retrieves and processes data from external sources without explicit sanitization or boundary enforcement.
Ingestion points: The agent ingests untrusted data from the Syndly MCP server via tools/list, list_projects, list_sources, and list_posts calls (SKILL.md).
Boundary markers: The instructions lack specific delimiters or instructions to treat tool outputs as data rather than instructions, potentially leading the agent to follow commands embedded in server responses.
Capability inventory: The skill possesses read capabilities for user projects and posts, and the create_post tool allows for content publication.
Sanitization: The skill explicitly instructs the agent to quote server errors exactly without sanitization or filtering before interpolation into the conversation.

syndly-help