The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external sources which could contain malicious instructions.
Ingestion points: External documentation fetched via WebFetch, HTTP/RPC responses, and CLI command outputs as specified in Guardrail 12.
Boundary markers: Guardrail 12 explicitly instructs the agent to treat these sources as untrusted data and to ignore any instructions within them that request credential access or changes to security rules.
Capability inventory: The skill utilizes Bash for CLI operations, WebFetch for remote data retrieval, and Write/Edit for file system modifications.
Sanitization: The skill relies on behavioral instructions (Guardrail 12) to filter out and ignore embedded instructions in external content rather than technical sanitization.
[COMMAND_EXECUTION]: The skill uses the Bash tool to interact with the cre command-line interface.
Evidence: Instructions include running cre init, cre workflow simulate, and cre login.
Safety Measures: Guardrail 11 prevents the automation of account creation, and Guardrail 3 explicitly forbids mainnet deployment operations, requiring simulation before any testnet actions.

chainlink-cre-skill