The Agent Skills Directory

[SAFE]: No security issues were detected. The skill provides legitimate instructions for using GitBook's developer tools and includes appropriate warnings regarding the handling of personal access tokens and secrets.
[PROMPT_INJECTION]: The skill describes building integrations that process external data, such as space content changes, which inherently introduces an indirect prompt injection surface. This is a functional aspect of the platform rather than a malicious instruction.
Ingestion points: Event handlers like space_content_updated as described in SKILL.md and references/runtime.md.
Boundary markers: None explicitly defined in the provided code templates for processing event data.
Capability inventory: The integration runtime provides access to context.api for authenticated platform operations and the standard fetch API for outgoing network requests.
Sanitization: No specific sanitization or validation techniques for event payloads are demonstrated in the instructional snippets.

build-integration