The Agent Skills Directory

[DATA_EXPOSURE_&_EXFILTRATION]: The skill instructions guide the agent to locate an API key (AGENT_READY_API_KEY) from the environment or .env files. This is a standard practice for secret management in developer tools and is implemented safely. The skill explicitly warns against passing secrets in URLs or query strings to prevent leakage in shell history, recommending the use of Authorization headers instead.
[REMOTE_CODE_EXECUTION]: The skill uses curl for API interactions and provides code examples in Python and TypeScript. All network operations are directed to the vendor's official domain (agent-ready.dev). No arbitrary or untrusted remote code is downloaded or executed.
[INDIRECT_PROMPT_INJECTION]: The skill acknowledges that scan results from external websites are untrusted data. It includes a specific security section advising the agent to treat response fields as inert data for summarization and not as instructions, which is a key mitigation for indirect prompt injection via scanned content.
[COMMAND_EXECUTION]: Shell commands provided (grep, curl, jq) are utility commands for local environment configuration and API interaction. There is no evidence of command injection or unsafe execution of user-controlled input.

agent-ready-api