The Agent Skills Directory

[SAFE]: The skill provides guidelines for adaptive orchestration of coding tasks using the Equilio framework. It prioritizes engineering principles, human review for consequential choices, and structured work artifacts. No executable scripts or suspicious commands are included.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data in the form of work goals, artifacts, and issue descriptions.
Ingestion points: Operations begin by reading existing work artifacts, project conventions, and goals as described in SKILL.md and references/work-artifact.md.
Boundary markers: The skill mandates a specific structure to isolate user-provided content. The references/work-artifact.md file requires preserving the 'Original Request' verbatim under a clear delimiter (---) and header, separating it from the agent's synthesized reasoning.
Capability inventory: The skill's primary capabilities involve reading/writing Markdown files and coordinating transitions between other specialized skills such as Frame, Map, Explore, Build, and Integrate.
Sanitization: The instructions focus on reconciling synthesized understanding against 'actual evidence' and 'verified behavior' to prevent hallucinations or adherence to invalid embedded instructions.
[EXTERNAL_DOWNLOADS]: The documentation references official technical writing and style guides from established services, including Google and Microsoft. These are used for developer orientation and do not represent a remote code execution risk.

equilio-loop