The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references documentation and community guidelines from well-known and trusted sources, including OpenAI, Anthropic, and Datadog, as well as established community repositories for agent skills.- [DATA_EXPOSURE]: The instructions emphasize data safety, specifically directing the agent to isolate provider credentials by owner, avoid destructive data migrations, and prevent the exposure of raw backend errors or stack traces in the public user interface.- [COMMAND_EXECUTION]: As part of a standard development workflow, the skill directs the agent to execute existing project scripts for testing, packaging, and verifying release assets. It advises using focused smoke tests to confirm configuration validity before deployment.- [INDIRECT_PROMPT_INJECTION]: While the skill interacts with project files (logs, data models, routes) that could theoretically contain malicious instructions, it mitigates risk by requiring review-first flows for AI-generated drafts and automated writeback actions.

app-feature-craft