The Agent Skills Directory

[DYNAMIC_CONTEXT_INJECTION]: The skill uses dynamic context markers (!) in SKILL.md to execute shell commands like git status, git branch, gh issue list, and gh auth status at load time. These commands are informational and used to provide context to the agent; they do not involve untrusted input or exfiltration.
[COMMAND_EXECUTION]: The skill instructs the agent to execute repository management commands, including gh issue create, gh label create, and quality validation tools such as pytest and npm test. These operations are well-scoped to the skill's primary function.
[INDIRECT_PROMPT_INJECTION]: The skill has a theoretical attack surface for indirect prompt injection as it processes external repository content.
Ingestion points: Reads .github/ISSUE_TEMPLATE/, CONTRIBUTING.md, and other repository guidelines mentioned in references/repository-templates.md.
Boundary markers: None; the skill does not specify delimiters to separate template content from system instructions.
Capability inventory: The skill has the capability to write to the repository via issue and label creation.
Sanitization: No sanitization or content validation is performed on the ingested repository files.

create-issues