The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool and assigns it to generated agents for tasks such as code formatting and running validation checks.
[EXTERNAL_DOWNLOADS]: The Research Engine component is designed to fetch official documentation and best practices using the Context7 MCP and tools like WebFetch or WebSearch when necessary.
[PROMPT_INJECTION]: As a generative skill, it possesses an attack surface for indirect prompt injection where user-provided requirements are interpolated into agent templates.
Ingestion points: User requirements enter the system through the Intelligence Engine (documented in SKILL.md and reference.md).
Boundary markers: Templates utilize structured YAML frontmatter and Markdown headers to delimit sections, although they lack explicit instructions to ignore embedded commands within variable substitutions.
Capability inventory: The skill and its generated agents have access to powerful tools including Bash, Write, Edit, and WebFetch (referenced in SKILL.md and examples.md).
Sanitization: The skill implements a 4-gate Validation Framework that checks YAML syntax, structure completeness, content quality, and TRUST 5 compliance (described in SKILL.md and reference.md).

moai-alfred-agent-factory