The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill promotes secure credential management by instructing users to read sensitive tokens from environment variables or external secret files (e.g., ~/.secrets/cf_access_token) instead of hardcoding them.
[COMMAND_EXECUTION]: Provides administrative command templates using standard tools like curl, jq, and dig to interact with Cloudflare's API and verify network status. These are appropriate for the skill's stated purpose of infrastructure configuration.
[EXTERNAL_DOWNLOADS]: Fetches network configuration data from Cloudflare's official API (api.cloudflare.com), which is a well-known service for this context.
[INDIRECT_PROMPT_INJECTION]: The skill features a data ingestion surface that processes external content.
Ingestion points: Fetches IP ranges from the Cloudflare API (https://api.cloudflare.com/client/v4/ips) as seen in references/origin-lockdown.md.
Boundary markers: None present in the command pipeline.
Capability inventory: Utilizes curl for network requests and jq for data processing within shell environments.
Sanitization: Relies on the integrity of the trusted API provider; no explicit sanitization is performed on the received JSON before processing.

build-cloudflare-access-sso