The Agent Skills Directory

[SAFE]: The skill implements standard authentication patterns using the @scalekit-sdk/node package. It follows security best practices by recommending HttpOnly, Secure, and SameSite cookie attributes to prevent common web vulnerabilities like XSS and CSRF.- [SAFE]: Sensitive credentials such as the Scalekit client secret are correctly handled using environment variables (process.env.SCALEKIT_CLIENT_SECRET) instead of hardcoding, adhering to secret management best practices.- [SAFE]: External links for documentation and supplemental resources target the vendor's official domains (docs.scalekit.com, mcp.scalekit.com), which are considered trusted within the context of this skill.- [INDIRECT_PROMPT_INJECTION]: The skill defines logic to process user-provided session tokens from cookies and authorization headers.
Ingestion points: Data enters the system via req.cookies?.accessToken and req.cookies?.refreshToken in SKILL.md.
Boundary markers: None explicitly defined in the prompt instructions for data separation.
Capability inventory: The skill initiates network requests to the Scalekit API via SDK methods in SKILL.md.
Sanitization: External tokens are passed to scalekit.validateAccessToken and scalekit.refreshAccessToken, which perform the necessary validation and parsing.

manage-saaskit-sessions