The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill facilitates the installation of development tools and libraries from the official vendor and well-known ecosystems. This includes the installation of @chainlink/contracts-ccip, @chainlink/local, @chainlink/ccip-sdk, and @chainlink/ccip-cli via npm, as well as OpenZeppelin contracts and forge-std via forge. These resources originate from trusted organizations and well-known technology providers.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external sources, creating a potential surface for indirect prompt injection. However, it incorporates robust defensive instructions to mitigate this risk.
Ingestion points: Data is fetched from docs.chain.link, api.ccip.chain.link, and various official GitHub repositories.
Boundary markers: Explicit instructions require the agent to treat all external output as untrusted and to ignore any embedded requests for secrets, shell execution, or changes to the skill's security guardrails.
Capability inventory: The agent utilizes WebFetch for live data retrieval and Write/Edit for generating contract and test files.
Sanitization: The skill mandates the omission of sensitive data solicitation and enforces that signing and transaction broadcasting must remain strictly within the user's manual control.

chainlink-ccip-skill