The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes content fetched from external sources, which could potentially contain malicious instructions intended to influence the agent's behavior.
Ingestion points: Data is retrieved through the heliusKnowledge tool via actions such as fetchHeliusBlog, searchSolanaDocs, getSIMD, and readSolanaSourceFile (specifically targeting the Helius blog, Solana documentation, and the Agave/Firedancer repositories).
Boundary markers: The skill instructions do not explicitly require the use of delimiters or 'ignore embedded instructions' warnings for the fetched content.
Capability inventory: The skill is restricted to information retrieval and synthesis; it does not possess capabilities for file-writing, system command execution, or unauthorized network communication.
Sanitization: There is no documented logic for sanitizing or validating the integrity of the external content before it is processed.
[EXTERNAL_DOWNLOADS]: The skill provides instructions for the user to manually install an external package if the required environment is not active.
Evidence: The SKILL.md file contains a recommendation for the user to run npx helius-mcp@2.2.0 to add the MCP server.
Context: The helius-mcp package is a resource provided by the skill's author (Helius Labs) and is expected for the skill's intended functionality.

svm