The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides instructions for downloading the CRE environment and templates from infrastructure hosted on the cre.chain.link domain and official GitHub repositories.
Evidence: Installation scripts referenced in getting-started.md at https://cre.chain.link/install.sh and https://cre.chain.link/install.ps1.
[REMOTE_CODE_EXECUTION]: The skill instructs users to install the CLI using piped shell commands from remote sources, which is a standard distribution method for this vendor's tooling.
Evidence: curl -sSfL https://cre.chain.link/install.sh | bash and irm https://cre.chain.link/install.ps1 | iex in getting-started.md.
[PROMPT_INJECTION]: The skill is designed to process external, potentially untrusted data from multiple sources, creating an indirect prompt injection attack surface.
Ingestion points: Data enters the agent's context through HTTP trigger payloads, EVM log event data (topics and data fields), and results from external API requests handled by the HTTP capability.
Boundary markers: The instructions mandate the use of Zod for runtime schema validation in TypeScript workflows to ensure external data matches expected structures.
Capability inventory: The agent utilizes Bash for CLI operations, WebFetch for fetching documentation, and Write/Edit for modifying project files and workflow code.
Sanitization: The skill emphasizes the use of scaled integers, bigints for blockchain values, and strict schema parsing to prevent malformed or malicious data from influencing execution logic.

chainlink-cre-skill