The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill utilizes the official NPM registry to install the GenLayer CLI package via npm install -g genlayer.
[COMMAND_EXECUTION]: The skill uses the Bash tool to perform legitimate development tasks, including local network management (genlayer up), account creation (genlayer account create), and contract interaction (genlayer call, genlayer write).
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it processes data from external sources such as transaction receipts, contract schema, and source code retrieved from the blockchain network.
Ingestion points: Data enters the context via genlayer receipt, genlayer call, genlayer schema, and genlayer code (SKILL.md).
Boundary markers: No explicit delimiters or warnings to ignore embedded instructions are provided in the instructions for these commands.
Capability inventory: The skill has broad capabilities to execute shell commands via the Bash tool (SKILL.md).
Sanitization: No sanitization or validation of the retrieved blockchain data is performed before it is presented to the agent.
[CREDENTIALS_UNSAFE]: The skill demonstrates non-interactive password entry for automation by piping strings to stdin (echo "password" | genlayer ...). While this is a common feature for CLI tools used in CI/CD environments, it is noted as a standard risk factor regarding shell command history.

genlayer-cli