The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external specifications to generate code and commands for implementation plans. This provides a surface for indirect prompt injection where a malicious requirement could influence the generated output.
Ingestion points: The skill processes a 'spec' or 'requirements' document provided as input to the writing task.
Boundary markers: The instructions do not define explicit boundary markers or 'ignore' instructions to isolate the input spec from the skill's operational logic.
Capability inventory: The skill generates plans containing functional Python code and shell commands (git, pytest) that are intended for immediate execution by subagents or via the executing-plans sub-skill.
Sanitization: There is no evidence of input validation, escaping, or sanitization for the content of the provided specifications.
[DYNAMIC_EXECUTION]: The skill generates implementation plans that include Python code snippets and shell scripts. These artifacts are explicitly designed to be executed via subagents or the executing-plans skill, representing a script generation and execution workflow.
[COMMAND_EXECUTION]: The task templates include direct shell command execution instructions, specifically using git for version control management and pytest for test execution within the local environment.

writing-plans