The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by ingesting external ticket descriptions to recommend monetization and retention strategies.
Ingestion points: User-provided tickets and the local data/conversion-rules.json data store.
Boundary markers: The skill defines a mandatory structured output format to constrain the agent's response.
Capability inventory: The skill orchestrates tasks by routing instructions to other agent skills, specifically vois-patterns for layout, righter for copy, and metrics-tagging for instrumentation.
Sanitization: The instructions do not specify any validation or sanitization steps for the external ticket data before it is processed by the agent.
[SAFE]: The Node.js script located in scripts/check-rule-sync.mjs is a benign utility for ensuring consistency between documentation and data.
The script relies exclusively on built-in Node.js modules (node:fs, node:url, path) to perform local file comparisons.
No network access or remote execution patterns were detected in the utility logic.

conversion-patterns