The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to download the Render CLI installation script from a remote source.
Evidence: curl -fsSL https://raw.githubusercontent.com/render-oss/cli/main/bin/install.sh in SKILL.md.
Context: The download targets a repository belonging to a well-known service (Render).
[REMOTE_CODE_EXECUTION]: The skill provides a command to pipe a remote shell script directly into the execution environment.
Evidence: curl -fsSL https://raw.githubusercontent.com/render-oss/cli/main/bin/install.sh | sh in SKILL.md.
Context: This is the official installation method for the Render CLI.
[COMMAND_EXECUTION]: The skill instructs the agent to execute various shell commands to verify the environment and validate infrastructure configurations.
Evidence: Commands include git remote -v, render --version, render blueprints validate, and render whoami -o json.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze project source code, which serves as an entry point for untrusted data that could contain malicious instructions.
Ingestion points: Analyzes project files (e.g., package.json, requirements.txt, go.mod) to detect frameworks and runtimes as defined in references/codebase-analysis.md.
Boundary markers: None identified to separate analyzed code from agent instructions.
Capability inventory: Includes file writing (render.yaml), shell command execution (Git and Render CLI), and network communication via the Render MCP server.
Sanitization: No explicit filtering or sanitization steps are defined for the data extracted from the user's codebase.

render-deploy