The Agent Skills Directory

[SAFE]: Detailed analysis of the skill's instructions and reference material confirms they are legitimate technical documentation. No malicious prompt injections, obfuscated payloads, unauthorized persistence mechanisms, or data exfiltration attempts were identified. The content is consistently scoped to teaching developers how to build durable applications with the Temporal SDK.- [EXTERNAL_DOWNLOADS]: The skill provides instructions for installing the Temporal CLI and various SDKs. All referenced URLs point to official vendor domains (temporal.download, github.com/temporalio) or standard, well-known public registries (PyPI, NPM). Third-party integrations mentioned (e.g., Google ADK, LangChain, Braintrust, Pydantic AI, and Vercel AI SDK) are legitimate industry projects, and references to them are documented neutrally for technical guidance.- [INDIRECT_PROMPT_INJECTION]: Several documentation files (e.g., ai-patterns.md) describe architectures for building AI agents that process untrusted user data. This is identified as a potential vulnerability surface for applications developed using these guides. However, the skill addresses this risk by recommending best practices such as structured data validation (Pydantic), centralized retry management, and the use of sandboxed activities for non-deterministic work.

temporal-developer