The Agent Skills Directory

[SAFE]: The skill functions as a static knowledge base and guidance tool. It does not perform network operations, execute shell commands, or access sensitive user data.
[INDIRECT_PROMPT_INJECTION]: The skill uses a local file system feedback loop to store and retrieve 'learnings' about the platform. While this creates a theoretical surface for indirect prompt injection, the risk is negligible due to the lack of dangerous capabilities in the skill.
Ingestion points: references/learnings.md and references/platform-guide.md are loaded into the agent context.
Boundary markers: No explicit delimiters are used to separate reference content from instructions.
Capability inventory: The skill performs no subprocess execution, network requests, or sensitive file operations.
Sanitization: No validation or escaping is applied to user-derived content before it is appended to the learnings file.

sales-painonsocial