The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references an external skill documentation at https://github.com/sickn33/agentic-awesome-skills/blob/main/skills/notion-manual-import/SKILL.md. This reference is used to guide users through the process of importing generated CSV data into Notion. The URL belongs to the vendor's own repository and is consistent with the skill's stated purpose of providing service register mappings.
[INDIRECT_PROMPT_INJECTION]: The skill ingests user-supplied information about business failures, technical stacks, and traffic patterns to generate structured outputs (SQL, JSON, CSV).
Ingestion points: User responses to the intake questionnaire in SKILL.md (Step 2).
Boundary markers: The instructions explicitly require the agent to treat unknown values as Unknown and forbid the invention of any data such as hostnames, error rates, or uptime figures.
Capability inventory: The skill has no tools enabled (tools: []). Its capabilities are limited to generating markdown and data artifacts within the chat context.
Sanitization: Instructions in references/security-safety-notes.md mandate that credentials, API keys, and sensitive customer data must never be pasted into the generated artifacts, effectively mitigating the risk of sensitive data interpolation.
[SAFE]: The skill incorporates comprehensive security best practices, including guidance on log scrubbing, monitoring from outside the infrastructure to avoid self-referential failure, and strictly defining alert severities to prevent alert fatigue. It emphasizes that the generated artifacts are for planning purposes and do not execute deployment or configuration changes.

observability-cloud-planning