The Agent Skills Directory

[SAFE]: The skill serves as a documentation guide for prompt engineering and does not contain executable malicious code or patterns.- [INDIRECT_PROMPT_INJECTION]: The skill provides templates (e.g., for Code Review and Content Writing) that use placeholders like [code] and [article text] to ingest external data.
Ingestion points: SKILL.md (template sections).
Boundary markers: Absent in the provided templates.
Capability inventory: Remote model execution via the belt CLI.
Sanitization: No sanitization or escaping mechanisms are described for the template placeholders.- [EXTERNAL_DOWNLOADS]: Recommends installing external tools and skills.
Recommends installation of belt-sh/cli via npx.
References installation instructions from github.com/inference-sh/skills.
Links to related skills such as inference-sh/skills@video-prompting-guide.

prompt-engineering