The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external, potentially untrusted data from multiple sources to influence its planning and file-writing operations.
Ingestion points: In SKILL.md (Step 1.2), the skill is instructed to read plan files, user-supplied files, task artifacts, and ticket or task descriptions.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore instructions' markers when processing content from these external sources.
Capability inventory: The skill possesses the capability to list directories using ls -La (Step 1.1), manage the git repository via git status and git checkout (Step 6.2 and 6.3), and modify local files (Step 4).
Sanitization: No explicit content sanitization or filtering logic is defined for the ingested data before it is integrated into the revised plan or used to determine next steps.
[COMMAND_EXECUTION]: The skill uses local shell and git commands to inspect the environment and manage the development workflow.
Evidence: The skill invokes ls -La (Step 1.1) for directory inspection and multiple git commands, including rev-parse, status, and checkout (Step 6.2 and 6.3), to synchronize the repository state.

iterate-plan