The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch instructions and executable files from an external repository (PostHog) and treat the retrieved content as part of the agent's core instructions for a task. This workflow introduces an indirect prompt injection surface where external data could potentially override or subvert the agent's intended behavior.
Ingestion points: The agent ingests external data through the posthog:skill-get tool (specifically the body field) and the posthog:skill-file-get tool (which retrieves bundled scripts and reference documents).
Boundary markers: The instructions do not define or suggest the use of boundary markers (like XML tags or delimiters) to isolate fetched content, nor do they include directives to ignore potential instructions embedded within the data.
Capability inventory: The skill enables the retrieval and execution of external scripts (e.g., Python code referenced in the scripts/ directory) and documentation that guides agent workflows.
Sanitization: The instructions lack any requirement for sanitizing, validating, or filtering the content retrieved from the PostHog store before it is interpreted or executed.

skills-store