The Agent Skills Directory

[SAFE]: The skill's primary function is to read internal repository documentation (GDDs, ADRs, and architecture manifests) and generate Markdown-based epic files. All operations are local to the repository environment.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it ingests and processes untrusted documentation files.
Ingestion points: Reads files from design/gdd/*.md, docs/architecture/*.md, and docs/engine-reference/ as specified in references/claude-skill.md.
Boundary markers: The instructions do not define explicit delimiters or instructions to ignore embedded commands within the processed GDD or ADR files.
Capability inventory: The skill uses Read, Write, Grep, and Task tools. The Write tool is restricted to creating Markdown files in the production/epics/ directory.
Sanitization: There is no explicit sanitization of the content extracted from GDDs before it is written into new EPIC.md files.
Context: Since the skill's primary purpose is document processing and its output is restricted to Markdown, the severity is minimal.

create-epics