The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent on how to build and execute the Avail node binary using common developer tools.
Evidence: Commands for building the project using cargo run --locked --release are provided in references/core-node-and-chains.md.
Evidence: Instructions for running the node in a containerized environment using docker run with volume mounting and network port exposure are documented in references/core-node-and-chains.md.
[INDIRECT_PROMPT_INJECTION]: The skill describes functionality for ingesting and processing data from an external, untrusted source (the Avail blockchain), which creates a potential surface for indirect prompt injection attacks.
Ingestion points: Data is ingested via various Kate RPC methods such as kate_queryRows, kate_queryProof, kate_queryMultiProof, and kate_queryDataProof as described in references/core-kate-rpc.md.
Boundary markers: No specific boundary markers or "ignore embedded instructions" warnings are included in the instructions for handling the data returned by these RPC calls.
Capability inventory: The node binary, when running, has the capability to perform network operations and local filesystem writes (to the specified data directory).
Sanitization: The skill does not provide specific instructions for sanitizing or validating the content retrieved from the blockchain before the agent processes it.

avail