The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands for project building and deployment automation. Specifically, it uses pnpm build to compile applications, zip to package source files, and curl to interact with the ibl.ai hosting API.
[DATA_EXFILTRATION]: The skill reads project source code and sensitive environment variables (from .env.local and iblai.env), packages them into a ZIP archive, and uploads them to the vendor's API endpoint (api.iblai.app). This is the core functionality of the deployment skill and is explicitly documented.
[CREDENTIALS_UNSAFE]: The skill accesses the TOKEN (platform API key) and IBLAI_API_KEY from local environment files to authorize deployment requests. It follows standard practices by advising users to manage these secrets in .env files.
[EXTERNAL_DOWNLOADS]: The skill references and fetches setup guidelines and documentation from the vendor's public GitHub repository at github.com/iblai/vibe.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the local environment, such as the project name from package.json and values from .env files, to configure deployment parameters. While it performs sanitization on the project slug, these ingestion points represent an attack surface for data-driven influence on the agent's behavior.

iblai-vibe-ops-deploy