The Agent Skills Directory

[SAFE]: The skill facilitates the discovery and management of agent workflows using official tools provided by PostHog, which is a well-known service. All remote interactions and data retrievals are performed against PostHog infrastructure.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest instruction sets and scripts from an external source (PostHog) and advises the agent to adopt retrieved content as system instructions. While this is the intended purpose of a skill store, it establishes a surface where agent behavior is guided by remote data.
Ingestion points: The posthog:skill-get tool retrieves a body field containing markdown instructions, and posthog:skill-file-get retrieves file content for scripts and reference documents.
Boundary markers: The skill does not define specific delimiters or "ignore embedded instructions" warnings for the retrieved content.
Capability inventory: The skill uses the PostHog MCP toolset, enabling the agent to list, fetch, create, and update remote skill assets.
Sanitization: No explicit sanitization or validation of the retrieved markdown instructions or script content is performed by the skill itself before implementation.

skills-store