The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill provides instructions to download and execute the official Render CLI installation script. This script is retrieved from the author's official GitHub repository and executed via a shell pipe.
Evidence: curl -fsSL https://raw.githubusercontent.com/render-oss/cli/main/bin/install.sh | sh in SKILL.md.
Context: The resource belongs to the render-oss organization, which corresponds to the skill's author.
[COMMAND_EXECUTION]: The skill performs various shell operations using the Render CLI and Git to manage deployments, validate configuration files, and retrieve account information.
Evidence: Commands such as render blueprints validate, render whoami, and git remote -v are used throughout the deployment workflow.
[EXTERNAL_DOWNLOADS]: The skill references external URLs for downloading documentation, templates, and the CLI tool from official Render domains.
Evidence: https://raw.githubusercontent.com/render-oss/cli/main/bin/install.sh and various dashboard.render.com links.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it reads and interprets data from local project files to automate configuration tasks.
Ingestion points: The agent reads local repository files such as package.json, requirements.txt, go.mod, and Pipfile to identify frameworks and build settings (described in SKILL.md and references/codebase-analysis.md).
Boundary markers: Absent; the agent is instructed to directly parse contents for analysis without explicit delimiters or instructions to ignore embedded commands.
Capability inventory: The skill has the ability to execute shell commands (render, git), write local files (render.yaml), and interact with the Render API via MCP tools.
Sanitization: Absent; values extracted from project files are used to generate deployment configurations and commands.

render-deploy