The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs the official CLI package @customerio/cli from the NPM registry and fetches various README files from the vendor's GitHub repositories (github.com/customerio/*) to guide integration and onboarding.
[COMMAND_EXECUTION]: Executes the cio CLI for account authentication, workspace discovery, domain configuration, and event tracking. It also invokes standard package managers (such as npm, pip, flutter, and go) to install SDKs in the user's project environment.
[PROMPT_INJECTION]: Identifies an indirect prompt injection surface where the agent is instructed to retrieve content from external vendor READMEs and execute instructions found within them.
Ingestion points: External README files fetched via curl in SKILL.md and integration.md, and API responses in billing.md and onboarding.md.
Boundary markers: None present for interpolated documentation content.
Capability inventory: Shell command execution (cio), package management (npm, pip, etc.), and local file system writes (.env).
Sanitization: No explicit sanitization or validation of the retrieved documentation content is performed before processing.
[CREDENTIALS_UNSAFE]: Handles sensitive service account tokens (sa_live_). The skill follows best practices by instructing the agent to pipe these tokens via stdin to avoid shell history logging and recommending the use of environment variables for application-level secret storage.
[DATA_EXFILTRATION]: Performs network operations to communicate with the vendor's official API endpoints (api.customer.io, api-eu.customer.io) for legitimate account management, billing checks, and data ingestion tasks.

cio