The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external graph datasets such as adjacency matrices and social network structures which could potentially contain malicious instructions embedded within node or edge data.
Ingestion points: Adjacency data passed to mcp__sublinear-time-solver__pageRank and mcp__sublinear-time-solver__analyzeMatrix in SKILL.md.
Capability inventory: The skill has the ability to execute code via mcp__flow-nexus__sandbox_execute.
Boundary markers: No explicit markers or instructions to ignore embedded content are provided in the ingestion logic.
Sanitization: No sanitization logic for graph data labels or properties is described.
[DYNAMIC_EXECUTION]: The skill demonstrates the generation and execution of Python code at runtime using the mcp__flow-nexus__sandbox_execute tool for distributed graph processing tasks as shown in the integration examples in SKILL.md.

agent-pagerank-analyzer