The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill instructs the user to install the CLI by piping a remote script directly into the shell. Because the source is the vendor's own repository, this is documented as standard distribution practice.
Evidence: curl -fsSL https://raw.githubusercontent.com/render-oss/cli/refs/heads/main/bin/install.sh | sh in SKILL.md.
[EXTERNAL_DOWNLOADS]: Fetches pre-compiled binary releases for the CLI from the vendor's GitHub repository for use in CI/CD pipelines.
Evidence: Download of cli_1.1.0_linux_amd64.zip from github.com/render-oss/cli in SKILL.md.
[PRIVILEGE_ESCALATION]: Documentation for CI/CD setup includes using sudo to move the downloaded binary into system execution paths.
Evidence: sudo mv cli_v1.1.0 /usr/local/bin/render in SKILL.md.
[COMMAND_EXECUTION]: The skill acts as a comprehensive reference for CLI commands that interact with cloud resources, including database sessions and remote terminal access.
Evidence: Documentation for render deploys, render psql, and render ssh in both SKILL.md and references/command-cheatsheet.md.
[INDIRECT_PROMPT_INJECTION]: The skill enables the agent to read external data from service logs and database queries, which could potentially contain adversarial instructions.
Ingestion points: Output from render logs and render psql commands.
Boundary markers: None specified in the instructions for isolating tool output.
Capability inventory: The skill provides capabilities for shell command execution and environment configuration.
Sanitization: No explicit sanitization of tool output is provided in the skill documentation.

render-cli