The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill's documentation provides instructions to download and execute a shell script from the vendor's repository (https://raw.githubusercontent.com/dwellir-public/cli/main/scripts/install.sh) by piping it directly to sh. This pattern executes remote code without local inspection.- [COMMAND_EXECUTION]: Multiple reference files contain examples of shell commands (e.g., 'dwellir endpoints search', 'dwellir docs get') that are intended to be executed in the agent's operating environment.- [EXTERNAL_DOWNLOADS]: The skill facilitates downloading the Dwellir CLI tool and fetching documentation from remote vendor-controlled sources.- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of data from external sources such as the Hyperliquid Info API and real-time data streams, creating an attack surface for instructions embedded in data.
Ingestion points: Market data queries in references/info-api.md and streaming updates in references/orderbook-websocket.md and references/grpc-gateway.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the prompt examples.
Capability inventory: The skill enables network operations (HTTPS, WSS, gRPC) and execution of the vendor's CLI tool.
Sanitization: The provided examples lack evidence of input validation or sanitization for data received from external API endpoints.

hyperliquid