The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes git log and git diff commands to analyze the project's history and current changes during its research phase.- [PROMPT_INJECTION]: The skill includes specific stylistic and behavioral instructions, such as 'ultrathink' and avoiding certain terminology ('newbies'). These are task-specific constraints and do not constitute an attempt to bypass safety guardrails.- [DATA_EXPOSURE]: The skill accesses the local codebase and git history to inform its planning. This data is processed locally and is not sent to any external endpoints or non-whitelisted domains.- [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes untrusted codebase content. Ingestion points: codebase search, git log, and git diff (File: SKILL.md). Boundary markers: Not present. Capability inventory: Writes files to .github/plans/ (File: SKILL.md). Sanitization: None specified for the ingested repository data.

create-local-plan