The Agent Skills Directory

[SAFE]: The skill instructs the agent to verify the existence of sensitive environment variables like MONGODB_URI in .env files. This is documented as a standard security audit practice and is necessary for the skill's primary function of troubleshooting database connections.
[COMMAND_EXECUTION]: The guide includes a verification script (scripts/test-mongodb-connection.ts) meant to be executed via shell (node or ts-node) to validate the database connection. This is a functional requirement for the skill's purpose.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes untrusted project files (e.g., ARCHITECTURE.md, .env, source code) while possessing the capability to execute shell commands and read files.
Ingestion points: Project configuration files (.env, .env.local), documentation (.agents/SYSTEM/ARCHITECTURE.md), and application source code.
Boundary markers: Absent; the skill does not explicitly instruct the agent to ignore embedded commands in the files it audits.
Capability inventory: File system read access and shell command execution (node, ts-node).
Sanitization: Absent; data from the project is used directly for auditing steps.

mongodb-atlas-checker